07_AMP_predictions_v01.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Making predictions on Protegrin-1 mutants "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here is the procedure used to predicty activity, selectivity and hemolysis in all possible single, double, and triple mutants of protegrin-1. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import torch\n",
    "import pickle\n",
    "import joblib\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib_venn import venn3, venn3_unweighted, venn2_unweighted\n",
    "from sklearn import metrics\n",
    "\n",
    "pd.set_option('display.max_columns', 100)\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load the pickled model trained only on protegrin variants \n",
    "\n",
    "# model trained on MBC data\n",
    "model_GNB_mbc = joblib.load('models/model_v01.2_GNB_esm2_15B_activity_classification_corr03.pkl')\n",
    "corr_columns_mic = pickle.load(open('models/correlated_columns_v01.2_esm2_15B_activity_corr03.pkl', 'rb'))\n",
    "\n",
    "# model trained on hemolysis data\n",
    "model_SVR_hemo = joblib.load('models/model_v01.2_Lasso_esm2_15B_Hemolysis_regression_corr06.pkl')\n",
    "corr_columns_hemo = pickle.load(open('models/correlated_columns_v01.2_esm2_15B_hemolysis_corr06.pkl', 'rb'))\n",
    "\n",
    "# model to predict selectivity\n",
    "model_SVR_sel = joblib.load('models/model_v01.2_SVR_esm2_15B_selectivity_regression_corr065.pkl')\n",
    "corr_columns_sel = pickle.load(open('models/correlated_columns_v01.2_esm2_15B_selectivity_corr065.pkl', 'rb'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Predictions on single mutation Protegrin-1 Variants"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>Sequence</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1mut_0</td>\n",
       "      <td>RGGRLCYCRRRFCVLVGR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1mut_1</td>\n",
       "      <td>RMGRLCYCRRRFCVCVGR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1mut_2</td>\n",
       "      <td>RGGRLCYCRRRFCVCVFR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1mut_3</td>\n",
       "      <td>EGGRLCYCRRRFCVCVGR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1mut_4</td>\n",
       "      <td>RGGNLCYCRRRFCVCVGR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>338</th>\n",
       "      <td>1mut_338</td>\n",
       "      <td>RGGRLCYSRRRFCVCVGR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>339</th>\n",
       "      <td>1mut_339</td>\n",
       "      <td>RGGRLCYCRRKFCVCVGR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>340</th>\n",
       "      <td>1mut_340</td>\n",
       "      <td>RGGRLCHCRRRFCVCVGR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>341</th>\n",
       "      <td>1mut_341</td>\n",
       "      <td>RGGRLCYCRRDFCVCVGR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>342</th>\n",
       "      <td>1mut_342</td>\n",
       "      <td>RGGRLCYCRRRFPVCVGR</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>343 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           ID            Sequence\n",
       "0      1mut_0  RGGRLCYCRRRFCVLVGR\n",
       "1      1mut_1  RMGRLCYCRRRFCVCVGR\n",
       "2      1mut_2  RGGRLCYCRRRFCVCVFR\n",
       "3      1mut_3  EGGRLCYCRRRFCVCVGR\n",
       "4      1mut_4  RGGNLCYCRRRFCVCVGR\n",
       "..        ...                 ...\n",
       "338  1mut_338  RGGRLCYSRRRFCVCVGR\n",
       "339  1mut_339  RGGRLCYCRRKFCVCVGR\n",
       "340  1mut_340  RGGRLCHCRRRFCVCVGR\n",
       "341  1mut_341  RGGRLCYCRRDFCVCVGR\n",
       "342  1mut_342  RGGRLCYCRRRFPVCVGR\n",
       "\n",
       "[343 rows x 2 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load meta_data with mutations on PG1\n",
    "meta_data = pd.read_csv('data/protegrin_variants_1mut.csv', index_col=0)\n",
    "meta_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "######################## Load the embedding ######################\n",
    "# since those embeddings are too large, I saved them in a separate file\n",
    "\n",
    "# dir_path = 'data/embedding/esm2_15B_pg1_1mut/'\n",
    "# embeddings = {}\n",
    "# for file in os.listdir(dir_path):\n",
    "#         file_path = str(dir_path) + file \n",
    "#         label = file.split('.pt')[0]\n",
    "#         embeddings[label] = np.array(torch.load(file_path)['mean_representations'][48])\n",
    "\n",
    "\n",
    "# # merge embeddings with metadata\n",
    "# df = meta_data.merge(pd.DataFrame.from_dict(embeddings, orient='index'), how='inner', left_on='ID', right_index=True)\n",
    "\n",
    "# df.to_pickle('data/embedding/esm2_15B_pg1_1mut.pkl')\n",
    "#df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>Sequence</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "      <th>17</th>\n",
       "      <th>18</th>\n",
       "      <th>19</th>\n",
       "      <th>20</th>\n",
       "      <th>21</th>\n",
       "      <th>22</th>\n",
       "      <th>23</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "      <th>26</th>\n",
       "      <th>27</th>\n",
       "      <th>28</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "      <th>33</th>\n",
       "      <th>34</th>\n",
       "      <th>35</th>\n",
       "      <th>36</th>\n",
       "      <th>37</th>\n",
       "      <th>38</th>\n",
       "      <th>39</th>\n",
       "      <th>40</th>\n",
       "      <th>41</th>\n",
       "      <th>42</th>\n",
       "      <th>43</th>\n",
       "      <th>44</th>\n",
       "      <th>45</th>\n",
       "      <th>46</th>\n",
       "      <th>47</th>\n",
       "      <th>...</th>\n",
       "      <th>5070</th>\n",
       "      <th>5071</th>\n",
       "      <th>5072</th>\n",
       "      <th>5073</th>\n",
       "      <th>5074</th>\n",
       "      <th>5075</th>\n",
       "      <th>5076</th>\n",
       "      <th>5077</th>\n",
       "      <th>5078</th>\n",
       "      <th>5079</th>\n",
       "      <th>5080</th>\n",
       "      <th>5081</th>\n",
       "      <th>5082</th>\n",
       "      <th>5083</th>\n",
       "      <th>5084</th>\n",
       "      <th>5085</th>\n",
       "      <th>5086</th>\n",
       "      <th>5087</th>\n",
       "      <th>5088</th>\n",
       "      <th>5089</th>\n",
       "      <th>5090</th>\n",
       "      <th>5091</th>\n",
       "      <th>5092</th>\n",
       "      <th>5093</th>\n",
       "      <th>5094</th>\n",
       "      <th>5095</th>\n",
       "      <th>5096</th>\n",
       "      <th>5097</th>\n",
       "      <th>5098</th>\n",
       "      <th>5099</th>\n",
       "      <th>5100</th>\n",
       "      <th>5101</th>\n",
       "      <th>5102</th>\n",
       "      <th>5103</th>\n",
       "      <th>5104</th>\n",
       "      <th>5105</th>\n",
       "      <th>5106</th>\n",
       "      <th>5107</th>\n",
       "      <th>5108</th>\n",
       "      <th>5109</th>\n",
       "      <th>5110</th>\n",
       "      <th>5111</th>\n",
       "      <th>5112</th>\n",
       "      <th>5113</th>\n",
       "      <th>5114</th>\n",
       "      <th>5115</th>\n",
       "      <th>5116</th>\n",
       "      <th>5117</th>\n",
       "      <th>5118</th>\n",
       "      <th>5119</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1mut_0</td>\n",
       "      <td>RGGRLCYCRRRFCVLVGR</td>\n",
       "      <td>0.139956</td>\n",
       "      <td>-0.125581</td>\n",
       "      <td>-0.076405</td>\n",
       "      <td>0.146426</td>\n",
       "      <td>-0.280027</td>\n",
       "      <td>0.177829</td>\n",
       "      <td>-0.103134</td>\n",
       "      <td>0.157826</td>\n",
       "      <td>-0.095869</td>\n",
       "      <td>0.238305</td>\n",
       "      <td>-0.166609</td>\n",
       "      <td>-0.106000</td>\n",
       "      <td>0.197645</td>\n",
       "      <td>0.147172</td>\n",
       "      <td>0.172095</td>\n",
       "      <td>-0.173993</td>\n",
       "      <td>0.049640</td>\n",
       "      <td>-0.142380</td>\n",
       "      <td>0.071343</td>\n",
       "      <td>-0.033491</td>\n",
       "      <td>0.175970</td>\n",
       "      <td>0.044800</td>\n",
       "      <td>-0.230723</td>\n",
       "      <td>0.116288</td>\n",
       "      <td>-0.228306</td>\n",
       "      <td>-0.045376</td>\n",
       "      <td>0.081696</td>\n",
       "      <td>-0.078286</td>\n",
       "      <td>0.017577</td>\n",
       "      <td>0.161617</td>\n",
       "      <td>-0.144766</td>\n",
       "      <td>-0.046883</td>\n",
       "      <td>-0.329819</td>\n",
       "      <td>-0.064292</td>\n",
       "      <td>-0.049258</td>\n",
       "      <td>0.030661</td>\n",
       "      <td>0.182056</td>\n",
       "      <td>-0.066748</td>\n",
       "      <td>0.331273</td>\n",
       "      <td>-0.024626</td>\n",
       "      <td>0.153560</td>\n",
       "      <td>-0.341233</td>\n",
       "      <td>0.098469</td>\n",
       "      <td>0.041766</td>\n",
       "      <td>-0.098336</td>\n",
       "      <td>0.335028</td>\n",
       "      <td>0.060487</td>\n",
       "      <td>0.080803</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.290288</td>\n",
       "      <td>-0.138646</td>\n",
       "      <td>0.337517</td>\n",
       "      <td>-0.036846</td>\n",
       "      <td>-0.145865</td>\n",
       "      <td>0.145739</td>\n",
       "      <td>-0.013954</td>\n",
       "      <td>-0.022534</td>\n",
       "      <td>0.266587</td>\n",
       "      <td>0.210630</td>\n",
       "      <td>-0.026933</td>\n",
       "      <td>0.026307</td>\n",
       "      <td>0.076589</td>\n",
       "      <td>0.187688</td>\n",
       "      <td>0.110229</td>\n",
       "      <td>-0.061802</td>\n",
       "      <td>-0.139272</td>\n",
       "      <td>0.225796</td>\n",
       "      <td>-0.113380</td>\n",
       "      <td>-0.024664</td>\n",
       "      <td>-0.230150</td>\n",
       "      <td>-0.086779</td>\n",
       "      <td>-0.018946</td>\n",
       "      <td>-0.150796</td>\n",
       "      <td>0.282690</td>\n",
       "      <td>0.062186</td>\n",
       "      <td>0.217737</td>\n",
       "      <td>0.001308</td>\n",
       "      <td>-0.076220</td>\n",
       "      <td>0.029293</td>\n",
       "      <td>0.134988</td>\n",
       "      <td>0.126751</td>\n",
       "      <td>-0.051001</td>\n",
       "      <td>-0.078870</td>\n",
       "      <td>-0.066645</td>\n",
       "      <td>-0.088527</td>\n",
       "      <td>-0.066581</td>\n",
       "      <td>-0.284267</td>\n",
       "      <td>-0.055163</td>\n",
       "      <td>0.033041</td>\n",
       "      <td>-0.139801</td>\n",
       "      <td>-0.113517</td>\n",
       "      <td>0.121989</td>\n",
       "      <td>-0.190335</td>\n",
       "      <td>0.056447</td>\n",
       "      <td>-0.192906</td>\n",
       "      <td>0.056169</td>\n",
       "      <td>-0.037111</td>\n",
       "      <td>-0.264767</td>\n",
       "      <td>0.061206</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1mut_1</td>\n",
       "      <td>RMGRLCYCRRRFCVCVGR</td>\n",
       "      <td>0.130926</td>\n",
       "      <td>0.000480</td>\n",
       "      <td>-0.139269</td>\n",
       "      <td>0.132204</td>\n",
       "      <td>-0.236016</td>\n",
       "      <td>0.171966</td>\n",
       "      <td>-0.085120</td>\n",
       "      <td>0.153231</td>\n",
       "      <td>-0.031941</td>\n",
       "      <td>0.153898</td>\n",
       "      <td>-0.142295</td>\n",
       "      <td>-0.114512</td>\n",
       "      <td>0.133251</td>\n",
       "      <td>0.141893</td>\n",
       "      <td>0.076887</td>\n",
       "      <td>-0.163610</td>\n",
       "      <td>0.135277</td>\n",
       "      <td>-0.086710</td>\n",
       "      <td>0.105006</td>\n",
       "      <td>0.039981</td>\n",
       "      <td>0.171417</td>\n",
       "      <td>-0.014597</td>\n",
       "      <td>-0.170309</td>\n",
       "      <td>0.056443</td>\n",
       "      <td>-0.123541</td>\n",
       "      <td>-0.021217</td>\n",
       "      <td>0.097796</td>\n",
       "      <td>-0.054952</td>\n",
       "      <td>0.091123</td>\n",
       "      <td>0.158605</td>\n",
       "      <td>-0.163539</td>\n",
       "      <td>-0.032665</td>\n",
       "      <td>-0.253657</td>\n",
       "      <td>-0.142816</td>\n",
       "      <td>-0.084620</td>\n",
       "      <td>0.004738</td>\n",
       "      <td>0.186130</td>\n",
       "      <td>-0.062229</td>\n",
       "      <td>0.138385</td>\n",
       "      <td>-0.051106</td>\n",
       "      <td>0.132185</td>\n",
       "      <td>-0.129907</td>\n",
       "      <td>0.131162</td>\n",
       "      <td>0.019662</td>\n",
       "      <td>-0.074399</td>\n",
       "      <td>0.299450</td>\n",
       "      <td>0.103770</td>\n",
       "      <td>0.101100</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.303995</td>\n",
       "      <td>-0.181548</td>\n",
       "      <td>0.189195</td>\n",
       "      <td>-0.035498</td>\n",
       "      <td>-0.048187</td>\n",
       "      <td>0.136608</td>\n",
       "      <td>-0.038978</td>\n",
       "      <td>0.003040</td>\n",
       "      <td>0.180409</td>\n",
       "      <td>0.252026</td>\n",
       "      <td>-0.034072</td>\n",
       "      <td>0.067309</td>\n",
       "      <td>0.028784</td>\n",
       "      <td>0.119834</td>\n",
       "      <td>0.231834</td>\n",
       "      <td>-0.008835</td>\n",
       "      <td>-0.155341</td>\n",
       "      <td>0.164639</td>\n",
       "      <td>-0.088200</td>\n",
       "      <td>-0.141192</td>\n",
       "      <td>-0.107459</td>\n",
       "      <td>-0.126025</td>\n",
       "      <td>-0.100561</td>\n",
       "      <td>-0.043595</td>\n",
       "      <td>0.288866</td>\n",
       "      <td>-0.020607</td>\n",
       "      <td>0.087057</td>\n",
       "      <td>-0.002016</td>\n",
       "      <td>-0.042150</td>\n",
       "      <td>0.094406</td>\n",
       "      <td>0.091099</td>\n",
       "      <td>0.142107</td>\n",
       "      <td>-0.021993</td>\n",
       "      <td>-0.063503</td>\n",
       "      <td>0.044875</td>\n",
       "      <td>0.009611</td>\n",
       "      <td>-0.126196</td>\n",
       "      <td>-0.259387</td>\n",
       "      <td>-0.037057</td>\n",
       "      <td>0.041199</td>\n",
       "      <td>-0.139124</td>\n",
       "      <td>-0.158397</td>\n",
       "      <td>0.033567</td>\n",
       "      <td>-0.222891</td>\n",
       "      <td>0.147657</td>\n",
       "      <td>-0.210249</td>\n",
       "      <td>-0.039008</td>\n",
       "      <td>-0.049254</td>\n",
       "      <td>-0.220128</td>\n",
       "      <td>0.026089</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1mut_2</td>\n",
       "      <td>RGGRLCYCRRRFCVCVFR</td>\n",
       "      <td>0.122494</td>\n",
       "      <td>0.015151</td>\n",
       "      <td>-0.113578</td>\n",
       "      <td>0.147190</td>\n",
       "      <td>-0.248815</td>\n",
       "      <td>0.153636</td>\n",
       "      <td>-0.082545</td>\n",
       "      <td>0.143023</td>\n",
       "      <td>-0.032666</td>\n",
       "      <td>0.198368</td>\n",
       "      <td>-0.141825</td>\n",
       "      <td>-0.072487</td>\n",
       "      <td>0.164319</td>\n",
       "      <td>0.155215</td>\n",
       "      <td>0.000999</td>\n",
       "      <td>-0.142618</td>\n",
       "      <td>0.028556</td>\n",
       "      <td>-0.091350</td>\n",
       "      <td>0.169577</td>\n",
       "      <td>-0.013866</td>\n",
       "      <td>0.158021</td>\n",
       "      <td>-0.034980</td>\n",
       "      <td>-0.236284</td>\n",
       "      <td>0.121333</td>\n",
       "      <td>-0.169987</td>\n",
       "      <td>0.009360</td>\n",
       "      <td>0.132464</td>\n",
       "      <td>-0.091888</td>\n",
       "      <td>0.103464</td>\n",
       "      <td>0.148294</td>\n",
       "      <td>-0.164126</td>\n",
       "      <td>-0.009009</td>\n",
       "      <td>-0.292603</td>\n",
       "      <td>-0.130214</td>\n",
       "      <td>-0.137328</td>\n",
       "      <td>0.002338</td>\n",
       "      <td>0.152275</td>\n",
       "      <td>-0.041424</td>\n",
       "      <td>0.209629</td>\n",
       "      <td>-0.034807</td>\n",
       "      <td>0.131335</td>\n",
       "      <td>-0.169956</td>\n",
       "      <td>0.116559</td>\n",
       "      <td>0.063441</td>\n",
       "      <td>-0.047805</td>\n",
       "      <td>0.292008</td>\n",
       "      <td>0.127569</td>\n",
       "      <td>0.090374</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.341078</td>\n",
       "      <td>-0.151316</td>\n",
       "      <td>0.166382</td>\n",
       "      <td>-0.083772</td>\n",
       "      <td>-0.031800</td>\n",
       "      <td>0.173639</td>\n",
       "      <td>-0.039404</td>\n",
       "      <td>-0.026641</td>\n",
       "      <td>0.183440</td>\n",
       "      <td>0.283105</td>\n",
       "      <td>-0.058492</td>\n",
       "      <td>0.033385</td>\n",
       "      <td>0.051223</td>\n",
       "      <td>0.157715</td>\n",
       "      <td>0.203213</td>\n",
       "      <td>-0.015537</td>\n",
       "      <td>-0.231012</td>\n",
       "      <td>0.176929</td>\n",
       "      <td>-0.108814</td>\n",
       "      <td>-0.151691</td>\n",
       "      <td>-0.155576</td>\n",
       "      <td>-0.138166</td>\n",
       "      <td>-0.069419</td>\n",
       "      <td>-0.081773</td>\n",
       "      <td>0.287196</td>\n",
       "      <td>-0.045575</td>\n",
       "      <td>0.097953</td>\n",
       "      <td>-0.033838</td>\n",
       "      <td>-0.036365</td>\n",
       "      <td>0.125570</td>\n",
       "      <td>0.049084</td>\n",
       "      <td>0.162091</td>\n",
       "      <td>0.003630</td>\n",
       "      <td>-0.094018</td>\n",
       "      <td>-0.003765</td>\n",
       "      <td>0.019716</td>\n",
       "      <td>-0.172657</td>\n",
       "      <td>-0.253969</td>\n",
       "      <td>-0.060248</td>\n",
       "      <td>0.032126</td>\n",
       "      <td>-0.131952</td>\n",
       "      <td>-0.182381</td>\n",
       "      <td>0.012476</td>\n",
       "      <td>-0.216764</td>\n",
       "      <td>0.082182</td>\n",
       "      <td>-0.167000</td>\n",
       "      <td>0.017850</td>\n",
       "      <td>-0.070759</td>\n",
       "      <td>-0.238455</td>\n",
       "      <td>0.011200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1mut_3</td>\n",
       "      <td>EGGRLCYCRRRFCVCVGR</td>\n",
       "      <td>0.137794</td>\n",
       "      <td>-0.026297</td>\n",
       "      <td>-0.100598</td>\n",
       "      <td>0.132454</td>\n",
       "      <td>-0.240978</td>\n",
       "      <td>0.123862</td>\n",
       "      <td>-0.108370</td>\n",
       "      <td>0.163518</td>\n",
       "      <td>-0.074148</td>\n",
       "      <td>0.135701</td>\n",
       "      <td>-0.147309</td>\n",
       "      <td>-0.161241</td>\n",
       "      <td>0.135352</td>\n",
       "      <td>0.173438</td>\n",
       "      <td>0.115235</td>\n",
       "      <td>-0.128381</td>\n",
       "      <td>0.062604</td>\n",
       "      <td>-0.100330</td>\n",
       "      <td>0.103954</td>\n",
       "      <td>0.043673</td>\n",
       "      <td>0.147754</td>\n",
       "      <td>0.030620</td>\n",
       "      <td>-0.217643</td>\n",
       "      <td>0.078211</td>\n",
       "      <td>-0.180088</td>\n",
       "      <td>-0.034668</td>\n",
       "      <td>0.076345</td>\n",
       "      <td>-0.030924</td>\n",
       "      <td>0.068764</td>\n",
       "      <td>0.133319</td>\n",
       "      <td>-0.132511</td>\n",
       "      <td>-0.038674</td>\n",
       "      <td>-0.252596</td>\n",
       "      <td>-0.127496</td>\n",
       "      <td>-0.098959</td>\n",
       "      <td>0.027856</td>\n",
       "      <td>0.196983</td>\n",
       "      <td>-0.067005</td>\n",
       "      <td>0.181675</td>\n",
       "      <td>-0.063434</td>\n",
       "      <td>0.197272</td>\n",
       "      <td>-0.178226</td>\n",
       "      <td>0.124807</td>\n",
       "      <td>0.080779</td>\n",
       "      <td>-0.057945</td>\n",
       "      <td>0.309352</td>\n",
       "      <td>0.085288</td>\n",
       "      <td>0.077937</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.238969</td>\n",
       "      <td>-0.153715</td>\n",
       "      <td>0.178666</td>\n",
       "      <td>-0.043892</td>\n",
       "      <td>-0.123913</td>\n",
       "      <td>0.146099</td>\n",
       "      <td>-0.052013</td>\n",
       "      <td>-0.003793</td>\n",
       "      <td>0.247806</td>\n",
       "      <td>0.245565</td>\n",
       "      <td>-0.039263</td>\n",
       "      <td>0.074070</td>\n",
       "      <td>0.054889</td>\n",
       "      <td>0.122570</td>\n",
       "      <td>0.174041</td>\n",
       "      <td>-0.029537</td>\n",
       "      <td>-0.161622</td>\n",
       "      <td>0.190393</td>\n",
       "      <td>-0.111680</td>\n",
       "      <td>-0.081653</td>\n",
       "      <td>-0.096576</td>\n",
       "      <td>-0.078513</td>\n",
       "      <td>-0.077252</td>\n",
       "      <td>-0.082461</td>\n",
       "      <td>0.261394</td>\n",
       "      <td>0.016504</td>\n",
       "      <td>0.065125</td>\n",
       "      <td>-0.017172</td>\n",
       "      <td>-0.042037</td>\n",
       "      <td>0.077371</td>\n",
       "      <td>0.101969</td>\n",
       "      <td>0.171801</td>\n",
       "      <td>0.012216</td>\n",
       "      <td>-0.091692</td>\n",
       "      <td>-0.028345</td>\n",
       "      <td>0.013726</td>\n",
       "      <td>-0.096583</td>\n",
       "      <td>-0.347500</td>\n",
       "      <td>-0.079666</td>\n",
       "      <td>0.017933</td>\n",
       "      <td>-0.133635</td>\n",
       "      <td>-0.175840</td>\n",
       "      <td>0.021177</td>\n",
       "      <td>-0.237654</td>\n",
       "      <td>0.146838</td>\n",
       "      <td>-0.213037</td>\n",
       "      <td>0.014747</td>\n",
       "      <td>-0.115953</td>\n",
       "      <td>-0.227792</td>\n",
       "      <td>0.040570</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1mut_4</td>\n",
       "      <td>RGGNLCYCRRRFCVCVGR</td>\n",
       "      <td>0.108533</td>\n",
       "      <td>-0.000612</td>\n",
       "      <td>-0.119274</td>\n",
       "      <td>0.135785</td>\n",
       "      <td>-0.260022</td>\n",
       "      <td>0.144817</td>\n",
       "      <td>-0.134771</td>\n",
       "      <td>0.222083</td>\n",
       "      <td>-0.071226</td>\n",
       "      <td>0.149381</td>\n",
       "      <td>-0.132229</td>\n",
       "      <td>-0.142084</td>\n",
       "      <td>0.202546</td>\n",
       "      <td>0.169355</td>\n",
       "      <td>0.116519</td>\n",
       "      <td>-0.167061</td>\n",
       "      <td>0.080885</td>\n",
       "      <td>-0.093418</td>\n",
       "      <td>0.138645</td>\n",
       "      <td>0.043515</td>\n",
       "      <td>0.134149</td>\n",
       "      <td>0.033941</td>\n",
       "      <td>-0.208214</td>\n",
       "      <td>0.081299</td>\n",
       "      <td>-0.160418</td>\n",
       "      <td>-0.035513</td>\n",
       "      <td>0.064047</td>\n",
       "      <td>-0.049382</td>\n",
       "      <td>0.048944</td>\n",
       "      <td>0.185777</td>\n",
       "      <td>-0.110163</td>\n",
       "      <td>-0.055136</td>\n",
       "      <td>-0.245780</td>\n",
       "      <td>-0.125597</td>\n",
       "      <td>-0.079749</td>\n",
       "      <td>0.033510</td>\n",
       "      <td>0.154440</td>\n",
       "      <td>-0.052611</td>\n",
       "      <td>0.233359</td>\n",
       "      <td>-0.057145</td>\n",
       "      <td>0.168876</td>\n",
       "      <td>-0.176145</td>\n",
       "      <td>0.141866</td>\n",
       "      <td>0.069709</td>\n",
       "      <td>-0.062727</td>\n",
       "      <td>0.325756</td>\n",
       "      <td>0.127122</td>\n",
       "      <td>0.116105</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.263345</td>\n",
       "      <td>-0.160914</td>\n",
       "      <td>0.203860</td>\n",
       "      <td>-0.063003</td>\n",
       "      <td>-0.075318</td>\n",
       "      <td>0.142121</td>\n",
       "      <td>0.016088</td>\n",
       "      <td>-0.019374</td>\n",
       "      <td>0.254149</td>\n",
       "      <td>0.238047</td>\n",
       "      <td>-0.070921</td>\n",
       "      <td>0.075543</td>\n",
       "      <td>0.025655</td>\n",
       "      <td>0.128815</td>\n",
       "      <td>0.166325</td>\n",
       "      <td>-0.076467</td>\n",
       "      <td>-0.168410</td>\n",
       "      <td>0.199253</td>\n",
       "      <td>-0.106218</td>\n",
       "      <td>-0.079883</td>\n",
       "      <td>-0.145478</td>\n",
       "      <td>-0.124827</td>\n",
       "      <td>-0.123468</td>\n",
       "      <td>-0.094550</td>\n",
       "      <td>0.282291</td>\n",
       "      <td>0.004593</td>\n",
       "      <td>0.082240</td>\n",
       "      <td>-0.030777</td>\n",
       "      <td>-0.038029</td>\n",
       "      <td>0.067017</td>\n",
       "      <td>0.078184</td>\n",
       "      <td>0.161283</td>\n",
       "      <td>-0.034093</td>\n",
       "      <td>-0.067665</td>\n",
       "      <td>-0.023683</td>\n",
       "      <td>-0.008662</td>\n",
       "      <td>-0.087251</td>\n",
       "      <td>-0.313730</td>\n",
       "      <td>-0.051508</td>\n",
       "      <td>0.005071</td>\n",
       "      <td>-0.155179</td>\n",
       "      <td>-0.159542</td>\n",
       "      <td>0.033873</td>\n",
       "      <td>-0.211086</td>\n",
       "      <td>0.123042</td>\n",
       "      <td>-0.241100</td>\n",
       "      <td>0.001314</td>\n",
       "      <td>-0.086826</td>\n",
       "      <td>-0.217587</td>\n",
       "      <td>0.040132</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>338</th>\n",
       "      <td>1mut_338</td>\n",
       "      <td>RGGRLCYSRRRFCVCVGR</td>\n",
       "      <td>0.171456</td>\n",
       "      <td>-0.061296</td>\n",
       "      <td>-0.072914</td>\n",
       "      <td>0.273927</td>\n",
       "      <td>-0.300305</td>\n",
       "      <td>0.117678</td>\n",
       "      <td>-0.101657</td>\n",
       "      <td>0.210263</td>\n",
       "      <td>-0.084550</td>\n",
       "      <td>0.291660</td>\n",
       "      <td>-0.135232</td>\n",
       "      <td>-0.103953</td>\n",
       "      <td>0.172183</td>\n",
       "      <td>0.159045</td>\n",
       "      <td>0.118081</td>\n",
       "      <td>-0.134593</td>\n",
       "      <td>-0.002135</td>\n",
       "      <td>-0.107870</td>\n",
       "      <td>0.084760</td>\n",
       "      <td>0.019784</td>\n",
       "      <td>0.110071</td>\n",
       "      <td>0.042066</td>\n",
       "      <td>-0.289126</td>\n",
       "      <td>0.063729</td>\n",
       "      <td>-0.269355</td>\n",
       "      <td>0.048923</td>\n",
       "      <td>0.082522</td>\n",
       "      <td>-0.105431</td>\n",
       "      <td>0.087310</td>\n",
       "      <td>0.168107</td>\n",
       "      <td>-0.133678</td>\n",
       "      <td>-0.096823</td>\n",
       "      <td>-0.306802</td>\n",
       "      <td>-0.068080</td>\n",
       "      <td>-0.133230</td>\n",
       "      <td>0.029162</td>\n",
       "      <td>0.175851</td>\n",
       "      <td>-0.108541</td>\n",
       "      <td>0.383502</td>\n",
       "      <td>0.002627</td>\n",
       "      <td>0.223638</td>\n",
       "      <td>-0.272455</td>\n",
       "      <td>0.100111</td>\n",
       "      <td>0.011716</td>\n",
       "      <td>-0.065513</td>\n",
       "      <td>0.277845</td>\n",
       "      <td>0.094852</td>\n",
       "      <td>0.099362</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.321741</td>\n",
       "      <td>-0.201256</td>\n",
       "      <td>0.263864</td>\n",
       "      <td>-0.045683</td>\n",
       "      <td>-0.053043</td>\n",
       "      <td>0.132980</td>\n",
       "      <td>-0.056922</td>\n",
       "      <td>0.008876</td>\n",
       "      <td>0.237592</td>\n",
       "      <td>0.226631</td>\n",
       "      <td>-0.030301</td>\n",
       "      <td>0.053813</td>\n",
       "      <td>0.082987</td>\n",
       "      <td>0.189639</td>\n",
       "      <td>0.077907</td>\n",
       "      <td>-0.007688</td>\n",
       "      <td>-0.217726</td>\n",
       "      <td>0.268569</td>\n",
       "      <td>-0.098079</td>\n",
       "      <td>0.037021</td>\n",
       "      <td>-0.225534</td>\n",
       "      <td>-0.057410</td>\n",
       "      <td>-0.062185</td>\n",
       "      <td>-0.145098</td>\n",
       "      <td>0.322035</td>\n",
       "      <td>0.081622</td>\n",
       "      <td>0.172986</td>\n",
       "      <td>-0.022200</td>\n",
       "      <td>-0.015189</td>\n",
       "      <td>0.098192</td>\n",
       "      <td>0.149023</td>\n",
       "      <td>0.118560</td>\n",
       "      <td>-0.052649</td>\n",
       "      <td>-0.121595</td>\n",
       "      <td>-0.139058</td>\n",
       "      <td>-0.076863</td>\n",
       "      <td>0.023087</td>\n",
       "      <td>-0.283470</td>\n",
       "      <td>-0.061043</td>\n",
       "      <td>-0.001715</td>\n",
       "      <td>-0.124384</td>\n",
       "      <td>-0.141402</td>\n",
       "      <td>0.084083</td>\n",
       "      <td>-0.126614</td>\n",
       "      <td>0.090866</td>\n",
       "      <td>-0.164546</td>\n",
       "      <td>0.101935</td>\n",
       "      <td>-0.093842</td>\n",
       "      <td>-0.241469</td>\n",
       "      <td>0.064410</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>339</th>\n",
       "      <td>1mut_339</td>\n",
       "      <td>RGGRLCYCRRKFCVCVGR</td>\n",
       "      <td>0.139664</td>\n",
       "      <td>-0.007477</td>\n",
       "      <td>-0.113494</td>\n",
       "      <td>0.125650</td>\n",
       "      <td>-0.257957</td>\n",
       "      <td>0.115235</td>\n",
       "      <td>-0.102949</td>\n",
       "      <td>0.165274</td>\n",
       "      <td>-0.037457</td>\n",
       "      <td>0.129628</td>\n",
       "      <td>-0.116410</td>\n",
       "      <td>-0.111243</td>\n",
       "      <td>0.152694</td>\n",
       "      <td>0.159444</td>\n",
       "      <td>0.108152</td>\n",
       "      <td>-0.137213</td>\n",
       "      <td>0.113646</td>\n",
       "      <td>-0.067449</td>\n",
       "      <td>0.114553</td>\n",
       "      <td>0.030258</td>\n",
       "      <td>0.124338</td>\n",
       "      <td>0.030815</td>\n",
       "      <td>-0.155588</td>\n",
       "      <td>0.072062</td>\n",
       "      <td>-0.142941</td>\n",
       "      <td>-0.050166</td>\n",
       "      <td>0.112694</td>\n",
       "      <td>-0.043422</td>\n",
       "      <td>0.019357</td>\n",
       "      <td>0.167160</td>\n",
       "      <td>-0.137914</td>\n",
       "      <td>-0.076533</td>\n",
       "      <td>-0.279363</td>\n",
       "      <td>-0.126060</td>\n",
       "      <td>-0.093979</td>\n",
       "      <td>0.032394</td>\n",
       "      <td>0.172499</td>\n",
       "      <td>-0.061792</td>\n",
       "      <td>0.175558</td>\n",
       "      <td>-0.087806</td>\n",
       "      <td>0.161482</td>\n",
       "      <td>-0.118832</td>\n",
       "      <td>0.130527</td>\n",
       "      <td>0.026239</td>\n",
       "      <td>-0.087232</td>\n",
       "      <td>0.309075</td>\n",
       "      <td>0.131613</td>\n",
       "      <td>0.099836</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.255017</td>\n",
       "      <td>-0.156844</td>\n",
       "      <td>0.188265</td>\n",
       "      <td>-0.047636</td>\n",
       "      <td>-0.010267</td>\n",
       "      <td>0.135643</td>\n",
       "      <td>0.006043</td>\n",
       "      <td>-0.046831</td>\n",
       "      <td>0.220550</td>\n",
       "      <td>0.258802</td>\n",
       "      <td>-0.059154</td>\n",
       "      <td>0.113412</td>\n",
       "      <td>0.003682</td>\n",
       "      <td>0.101039</td>\n",
       "      <td>0.176312</td>\n",
       "      <td>-0.023857</td>\n",
       "      <td>-0.145524</td>\n",
       "      <td>0.163611</td>\n",
       "      <td>-0.092816</td>\n",
       "      <td>-0.087014</td>\n",
       "      <td>-0.064904</td>\n",
       "      <td>-0.182796</td>\n",
       "      <td>-0.146189</td>\n",
       "      <td>-0.044013</td>\n",
       "      <td>0.285041</td>\n",
       "      <td>0.000105</td>\n",
       "      <td>0.048911</td>\n",
       "      <td>-0.003084</td>\n",
       "      <td>-0.068871</td>\n",
       "      <td>0.079498</td>\n",
       "      <td>0.110505</td>\n",
       "      <td>0.126899</td>\n",
       "      <td>0.010283</td>\n",
       "      <td>-0.083637</td>\n",
       "      <td>0.003246</td>\n",
       "      <td>0.031959</td>\n",
       "      <td>-0.092911</td>\n",
       "      <td>-0.287391</td>\n",
       "      <td>-0.017080</td>\n",
       "      <td>0.023965</td>\n",
       "      <td>-0.184101</td>\n",
       "      <td>-0.144901</td>\n",
       "      <td>0.060435</td>\n",
       "      <td>-0.246207</td>\n",
       "      <td>0.152064</td>\n",
       "      <td>-0.207770</td>\n",
       "      <td>-0.011471</td>\n",
       "      <td>-0.131945</td>\n",
       "      <td>-0.231710</td>\n",
       "      <td>0.057206</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>340</th>\n",
       "      <td>1mut_340</td>\n",
       "      <td>RGGRLCHCRRRFCVCVGR</td>\n",
       "      <td>0.153120</td>\n",
       "      <td>-0.080120</td>\n",
       "      <td>-0.124630</td>\n",
       "      <td>0.133538</td>\n",
       "      <td>-0.255544</td>\n",
       "      <td>0.160656</td>\n",
       "      <td>-0.103168</td>\n",
       "      <td>0.154334</td>\n",
       "      <td>-0.048304</td>\n",
       "      <td>0.185473</td>\n",
       "      <td>-0.124932</td>\n",
       "      <td>-0.117904</td>\n",
       "      <td>0.179380</td>\n",
       "      <td>0.147361</td>\n",
       "      <td>0.100039</td>\n",
       "      <td>-0.155124</td>\n",
       "      <td>0.040069</td>\n",
       "      <td>-0.122986</td>\n",
       "      <td>0.153903</td>\n",
       "      <td>-0.003839</td>\n",
       "      <td>0.148773</td>\n",
       "      <td>0.025067</td>\n",
       "      <td>-0.242416</td>\n",
       "      <td>0.108469</td>\n",
       "      <td>-0.166943</td>\n",
       "      <td>-0.020389</td>\n",
       "      <td>0.062911</td>\n",
       "      <td>-0.067076</td>\n",
       "      <td>0.057570</td>\n",
       "      <td>0.160443</td>\n",
       "      <td>-0.158911</td>\n",
       "      <td>-0.034387</td>\n",
       "      <td>-0.332806</td>\n",
       "      <td>-0.126596</td>\n",
       "      <td>-0.082752</td>\n",
       "      <td>0.011322</td>\n",
       "      <td>0.174122</td>\n",
       "      <td>-0.069177</td>\n",
       "      <td>0.254445</td>\n",
       "      <td>-0.028243</td>\n",
       "      <td>0.128014</td>\n",
       "      <td>-0.191003</td>\n",
       "      <td>0.104212</td>\n",
       "      <td>0.113486</td>\n",
       "      <td>-0.089644</td>\n",
       "      <td>0.287929</td>\n",
       "      <td>0.059788</td>\n",
       "      <td>0.115429</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.308235</td>\n",
       "      <td>-0.154275</td>\n",
       "      <td>0.266652</td>\n",
       "      <td>-0.090237</td>\n",
       "      <td>-0.035987</td>\n",
       "      <td>0.118418</td>\n",
       "      <td>-0.051667</td>\n",
       "      <td>0.000337</td>\n",
       "      <td>0.229528</td>\n",
       "      <td>0.222554</td>\n",
       "      <td>-0.052118</td>\n",
       "      <td>0.005027</td>\n",
       "      <td>0.032014</td>\n",
       "      <td>0.162373</td>\n",
       "      <td>0.125081</td>\n",
       "      <td>-0.104945</td>\n",
       "      <td>-0.201631</td>\n",
       "      <td>0.208292</td>\n",
       "      <td>-0.101363</td>\n",
       "      <td>-0.047418</td>\n",
       "      <td>-0.151285</td>\n",
       "      <td>-0.122030</td>\n",
       "      <td>-0.051473</td>\n",
       "      <td>-0.128324</td>\n",
       "      <td>0.273887</td>\n",
       "      <td>0.032952</td>\n",
       "      <td>0.174311</td>\n",
       "      <td>-0.035650</td>\n",
       "      <td>-0.026444</td>\n",
       "      <td>0.056570</td>\n",
       "      <td>0.084502</td>\n",
       "      <td>0.140714</td>\n",
       "      <td>-0.054317</td>\n",
       "      <td>-0.064304</td>\n",
       "      <td>-0.016965</td>\n",
       "      <td>-0.037114</td>\n",
       "      <td>-0.067974</td>\n",
       "      <td>-0.293219</td>\n",
       "      <td>-0.059548</td>\n",
       "      <td>0.028274</td>\n",
       "      <td>-0.170386</td>\n",
       "      <td>-0.156650</td>\n",
       "      <td>0.046467</td>\n",
       "      <td>-0.211848</td>\n",
       "      <td>0.081548</td>\n",
       "      <td>-0.178534</td>\n",
       "      <td>0.055623</td>\n",
       "      <td>-0.080532</td>\n",
       "      <td>-0.263759</td>\n",
       "      <td>0.021452</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>341</th>\n",
       "      <td>1mut_341</td>\n",
       "      <td>RGGRLCYCRRDFCVCVGR</td>\n",
       "      <td>0.140428</td>\n",
       "      <td>0.020919</td>\n",
       "      <td>-0.108401</td>\n",
       "      <td>0.110415</td>\n",
       "      <td>-0.265068</td>\n",
       "      <td>0.067399</td>\n",
       "      <td>-0.120239</td>\n",
       "      <td>0.202811</td>\n",
       "      <td>-0.071523</td>\n",
       "      <td>0.135932</td>\n",
       "      <td>-0.115880</td>\n",
       "      <td>-0.131030</td>\n",
       "      <td>0.170962</td>\n",
       "      <td>0.215204</td>\n",
       "      <td>0.116121</td>\n",
       "      <td>-0.101719</td>\n",
       "      <td>0.099168</td>\n",
       "      <td>-0.089447</td>\n",
       "      <td>0.148354</td>\n",
       "      <td>0.085804</td>\n",
       "      <td>0.143292</td>\n",
       "      <td>0.013764</td>\n",
       "      <td>-0.173598</td>\n",
       "      <td>0.075202</td>\n",
       "      <td>-0.153304</td>\n",
       "      <td>-0.056858</td>\n",
       "      <td>0.067892</td>\n",
       "      <td>-0.008968</td>\n",
       "      <td>0.037887</td>\n",
       "      <td>0.132698</td>\n",
       "      <td>-0.131542</td>\n",
       "      <td>-0.042660</td>\n",
       "      <td>-0.218446</td>\n",
       "      <td>-0.116871</td>\n",
       "      <td>-0.071132</td>\n",
       "      <td>0.008414</td>\n",
       "      <td>0.185795</td>\n",
       "      <td>-0.082029</td>\n",
       "      <td>0.175870</td>\n",
       "      <td>-0.093090</td>\n",
       "      <td>0.199076</td>\n",
       "      <td>-0.139710</td>\n",
       "      <td>0.115822</td>\n",
       "      <td>0.061500</td>\n",
       "      <td>-0.093693</td>\n",
       "      <td>0.303419</td>\n",
       "      <td>0.101090</td>\n",
       "      <td>0.110945</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.247683</td>\n",
       "      <td>-0.131133</td>\n",
       "      <td>0.114285</td>\n",
       "      <td>-0.049539</td>\n",
       "      <td>-0.046005</td>\n",
       "      <td>0.125978</td>\n",
       "      <td>-0.016459</td>\n",
       "      <td>-0.006963</td>\n",
       "      <td>0.209881</td>\n",
       "      <td>0.270406</td>\n",
       "      <td>-0.059499</td>\n",
       "      <td>0.091360</td>\n",
       "      <td>0.007245</td>\n",
       "      <td>0.104477</td>\n",
       "      <td>0.172886</td>\n",
       "      <td>-0.041010</td>\n",
       "      <td>-0.134691</td>\n",
       "      <td>0.171602</td>\n",
       "      <td>-0.101408</td>\n",
       "      <td>-0.083674</td>\n",
       "      <td>-0.061620</td>\n",
       "      <td>-0.128539</td>\n",
       "      <td>-0.090669</td>\n",
       "      <td>-0.067588</td>\n",
       "      <td>0.277691</td>\n",
       "      <td>-0.028381</td>\n",
       "      <td>0.046004</td>\n",
       "      <td>-0.016129</td>\n",
       "      <td>-0.058107</td>\n",
       "      <td>0.096573</td>\n",
       "      <td>0.124325</td>\n",
       "      <td>0.192702</td>\n",
       "      <td>0.022009</td>\n",
       "      <td>-0.109851</td>\n",
       "      <td>0.015895</td>\n",
       "      <td>0.049376</td>\n",
       "      <td>-0.090216</td>\n",
       "      <td>-0.352958</td>\n",
       "      <td>-0.018720</td>\n",
       "      <td>0.004421</td>\n",
       "      <td>-0.178036</td>\n",
       "      <td>-0.131529</td>\n",
       "      <td>0.062716</td>\n",
       "      <td>-0.249751</td>\n",
       "      <td>0.126609</td>\n",
       "      <td>-0.227497</td>\n",
       "      <td>-0.025272</td>\n",
       "      <td>-0.121043</td>\n",
       "      <td>-0.241836</td>\n",
       "      <td>0.042768</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>342</th>\n",
       "      <td>1mut_342</td>\n",
       "      <td>RGGRLCYCRRRFPVCVGR</td>\n",
       "      <td>0.119557</td>\n",
       "      <td>-0.117891</td>\n",
       "      <td>-0.044751</td>\n",
       "      <td>0.223649</td>\n",
       "      <td>-0.225591</td>\n",
       "      <td>0.184604</td>\n",
       "      <td>-0.120514</td>\n",
       "      <td>0.216192</td>\n",
       "      <td>-0.127079</td>\n",
       "      <td>0.331392</td>\n",
       "      <td>-0.135417</td>\n",
       "      <td>-0.049815</td>\n",
       "      <td>0.179718</td>\n",
       "      <td>0.153649</td>\n",
       "      <td>0.186285</td>\n",
       "      <td>-0.172529</td>\n",
       "      <td>0.002667</td>\n",
       "      <td>-0.112276</td>\n",
       "      <td>0.124021</td>\n",
       "      <td>-0.021766</td>\n",
       "      <td>0.086690</td>\n",
       "      <td>0.066704</td>\n",
       "      <td>-0.268331</td>\n",
       "      <td>0.125196</td>\n",
       "      <td>-0.246302</td>\n",
       "      <td>0.014953</td>\n",
       "      <td>0.089700</td>\n",
       "      <td>-0.086845</td>\n",
       "      <td>-0.000385</td>\n",
       "      <td>0.156274</td>\n",
       "      <td>-0.142530</td>\n",
       "      <td>-0.045533</td>\n",
       "      <td>-0.338137</td>\n",
       "      <td>-0.070790</td>\n",
       "      <td>-0.012801</td>\n",
       "      <td>0.067026</td>\n",
       "      <td>0.214020</td>\n",
       "      <td>-0.076741</td>\n",
       "      <td>0.344897</td>\n",
       "      <td>-0.012566</td>\n",
       "      <td>0.126707</td>\n",
       "      <td>-0.309104</td>\n",
       "      <td>0.137018</td>\n",
       "      <td>0.024018</td>\n",
       "      <td>-0.095259</td>\n",
       "      <td>0.323033</td>\n",
       "      <td>0.036483</td>\n",
       "      <td>0.085557</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.350772</td>\n",
       "      <td>-0.232596</td>\n",
       "      <td>0.292454</td>\n",
       "      <td>-0.092167</td>\n",
       "      <td>-0.047533</td>\n",
       "      <td>0.123823</td>\n",
       "      <td>-0.004531</td>\n",
       "      <td>0.018452</td>\n",
       "      <td>0.234012</td>\n",
       "      <td>0.212497</td>\n",
       "      <td>-0.085228</td>\n",
       "      <td>0.007345</td>\n",
       "      <td>0.057953</td>\n",
       "      <td>0.194741</td>\n",
       "      <td>0.074880</td>\n",
       "      <td>-0.031264</td>\n",
       "      <td>-0.179551</td>\n",
       "      <td>0.225836</td>\n",
       "      <td>-0.087978</td>\n",
       "      <td>0.025476</td>\n",
       "      <td>-0.189744</td>\n",
       "      <td>-0.065764</td>\n",
       "      <td>-0.042931</td>\n",
       "      <td>-0.169635</td>\n",
       "      <td>0.280548</td>\n",
       "      <td>0.072033</td>\n",
       "      <td>0.232753</td>\n",
       "      <td>-0.033540</td>\n",
       "      <td>0.023878</td>\n",
       "      <td>0.049674</td>\n",
       "      <td>0.110939</td>\n",
       "      <td>0.091594</td>\n",
       "      <td>-0.054995</td>\n",
       "      <td>-0.092958</td>\n",
       "      <td>-0.159015</td>\n",
       "      <td>-0.081230</td>\n",
       "      <td>0.003357</td>\n",
       "      <td>-0.259860</td>\n",
       "      <td>-0.099152</td>\n",
       "      <td>0.050142</td>\n",
       "      <td>-0.154672</td>\n",
       "      <td>-0.116942</td>\n",
       "      <td>0.155337</td>\n",
       "      <td>-0.169825</td>\n",
       "      <td>0.113419</td>\n",
       "      <td>-0.165286</td>\n",
       "      <td>0.070795</td>\n",
       "      <td>0.003111</td>\n",
       "      <td>-0.265938</td>\n",
       "      <td>-0.002164</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>343 rows × 5122 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           ID            Sequence         0         1         2         3  \\\n",
       "0      1mut_0  RGGRLCYCRRRFCVLVGR  0.139956 -0.125581 -0.076405  0.146426   \n",
       "1      1mut_1  RMGRLCYCRRRFCVCVGR  0.130926  0.000480 -0.139269  0.132204   \n",
       "2      1mut_2  RGGRLCYCRRRFCVCVFR  0.122494  0.015151 -0.113578  0.147190   \n",
       "3      1mut_3  EGGRLCYCRRRFCVCVGR  0.137794 -0.026297 -0.100598  0.132454   \n",
       "4      1mut_4  RGGNLCYCRRRFCVCVGR  0.108533 -0.000612 -0.119274  0.135785   \n",
       "..        ...                 ...       ...       ...       ...       ...   \n",
       "338  1mut_338  RGGRLCYSRRRFCVCVGR  0.171456 -0.061296 -0.072914  0.273927   \n",
       "339  1mut_339  RGGRLCYCRRKFCVCVGR  0.139664 -0.007477 -0.113494  0.125650   \n",
       "340  1mut_340  RGGRLCHCRRRFCVCVGR  0.153120 -0.080120 -0.124630  0.133538   \n",
       "341  1mut_341  RGGRLCYCRRDFCVCVGR  0.140428  0.020919 -0.108401  0.110415   \n",
       "342  1mut_342  RGGRLCYCRRRFPVCVGR  0.119557 -0.117891 -0.044751  0.223649   \n",
       "\n",
       "            4         5         6         7         8         9        10  \\\n",
       "0   -0.280027  0.177829 -0.103134  0.157826 -0.095869  0.238305 -0.166609   \n",
       "1   -0.236016  0.171966 -0.085120  0.153231 -0.031941  0.153898 -0.142295   \n",
       "2   -0.248815  0.153636 -0.082545  0.143023 -0.032666  0.198368 -0.141825   \n",
       "3   -0.240978  0.123862 -0.108370  0.163518 -0.074148  0.135701 -0.147309   \n",
       "4   -0.260022  0.144817 -0.134771  0.222083 -0.071226  0.149381 -0.132229   \n",
       "..        ...       ...       ...       ...       ...       ...       ...   \n",
       "338 -0.300305  0.117678 -0.101657  0.210263 -0.084550  0.291660 -0.135232   \n",
       "339 -0.257957  0.115235 -0.102949  0.165274 -0.037457  0.129628 -0.116410   \n",
       "340 -0.255544  0.160656 -0.103168  0.154334 -0.048304  0.185473 -0.124932   \n",
       "341 -0.265068  0.067399 -0.120239  0.202811 -0.071523  0.135932 -0.115880   \n",
       "342 -0.225591  0.184604 -0.120514  0.216192 -0.127079  0.331392 -0.135417   \n",
       "\n",
       "           11        12        13        14        15        16        17  \\\n",
       "0   -0.106000  0.197645  0.147172  0.172095 -0.173993  0.049640 -0.142380   \n",
       "1   -0.114512  0.133251  0.141893  0.076887 -0.163610  0.135277 -0.086710   \n",
       "2   -0.072487  0.164319  0.155215  0.000999 -0.142618  0.028556 -0.091350   \n",
       "3   -0.161241  0.135352  0.173438  0.115235 -0.128381  0.062604 -0.100330   \n",
       "4   -0.142084  0.202546  0.169355  0.116519 -0.167061  0.080885 -0.093418   \n",
       "..        ...       ...       ...       ...       ...       ...       ...   \n",
       "338 -0.103953  0.172183  0.159045  0.118081 -0.134593 -0.002135 -0.107870   \n",
       "339 -0.111243  0.152694  0.159444  0.108152 -0.137213  0.113646 -0.067449   \n",
       "340 -0.117904  0.179380  0.147361  0.100039 -0.155124  0.040069 -0.122986   \n",
       "341 -0.131030  0.170962  0.215204  0.116121 -0.101719  0.099168 -0.089447   \n",
       "342 -0.049815  0.179718  0.153649  0.186285 -0.172529  0.002667 -0.112276   \n",
       "\n",
       "           18        19        20        21        22        23        24  \\\n",
       "0    0.071343 -0.033491  0.175970  0.044800 -0.230723  0.116288 -0.228306   \n",
       "1    0.105006  0.039981  0.171417 -0.014597 -0.170309  0.056443 -0.123541   \n",
       "2    0.169577 -0.013866  0.158021 -0.034980 -0.236284  0.121333 -0.169987   \n",
       "3    0.103954  0.043673  0.147754  0.030620 -0.217643  0.078211 -0.180088   \n",
       "4    0.138645  0.043515  0.134149  0.033941 -0.208214  0.081299 -0.160418   \n",
       "..        ...       ...       ...       ...       ...       ...       ...   \n",
       "338  0.084760  0.019784  0.110071  0.042066 -0.289126  0.063729 -0.269355   \n",
       "339  0.114553  0.030258  0.124338  0.030815 -0.155588  0.072062 -0.142941   \n",
       "340  0.153903 -0.003839  0.148773  0.025067 -0.242416  0.108469 -0.166943   \n",
       "341  0.148354  0.085804  0.143292  0.013764 -0.173598  0.075202 -0.153304   \n",
       "342  0.124021 -0.021766  0.086690  0.066704 -0.268331  0.125196 -0.246302   \n",
       "\n",
       "           25        26        27        28        29        30        31  \\\n",
       "0   -0.045376  0.081696 -0.078286  0.017577  0.161617 -0.144766 -0.046883   \n",
       "1   -0.021217  0.097796 -0.054952  0.091123  0.158605 -0.163539 -0.032665   \n",
       "2    0.009360  0.132464 -0.091888  0.103464  0.148294 -0.164126 -0.009009   \n",
       "3   -0.034668  0.076345 -0.030924  0.068764  0.133319 -0.132511 -0.038674   \n",
       "4   -0.035513  0.064047 -0.049382  0.048944  0.185777 -0.110163 -0.055136   \n",
       "..        ...       ...       ...       ...       ...       ...       ...   \n",
       "338  0.048923  0.082522 -0.105431  0.087310  0.168107 -0.133678 -0.096823   \n",
       "339 -0.050166  0.112694 -0.043422  0.019357  0.167160 -0.137914 -0.076533   \n",
       "340 -0.020389  0.062911 -0.067076  0.057570  0.160443 -0.158911 -0.034387   \n",
       "341 -0.056858  0.067892 -0.008968  0.037887  0.132698 -0.131542 -0.042660   \n",
       "342  0.014953  0.089700 -0.086845 -0.000385  0.156274 -0.142530 -0.045533   \n",
       "\n",
       "           32        33        34        35        36        37        38  \\\n",
       "0   -0.329819 -0.064292 -0.049258  0.030661  0.182056 -0.066748  0.331273   \n",
       "1   -0.253657 -0.142816 -0.084620  0.004738  0.186130 -0.062229  0.138385   \n",
       "2   -0.292603 -0.130214 -0.137328  0.002338  0.152275 -0.041424  0.209629   \n",
       "3   -0.252596 -0.127496 -0.098959  0.027856  0.196983 -0.067005  0.181675   \n",
       "4   -0.245780 -0.125597 -0.079749  0.033510  0.154440 -0.052611  0.233359   \n",
       "..        ...       ...       ...       ...       ...       ...       ...   \n",
       "338 -0.306802 -0.068080 -0.133230  0.029162  0.175851 -0.108541  0.383502   \n",
       "339 -0.279363 -0.126060 -0.093979  0.032394  0.172499 -0.061792  0.175558   \n",
       "340 -0.332806 -0.126596 -0.082752  0.011322  0.174122 -0.069177  0.254445   \n",
       "341 -0.218446 -0.116871 -0.071132  0.008414  0.185795 -0.082029  0.175870   \n",
       "342 -0.338137 -0.070790 -0.012801  0.067026  0.214020 -0.076741  0.344897   \n",
       "\n",
       "           39        40        41        42        43        44        45  \\\n",
       "0   -0.024626  0.153560 -0.341233  0.098469  0.041766 -0.098336  0.335028   \n",
       "1   -0.051106  0.132185 -0.129907  0.131162  0.019662 -0.074399  0.299450   \n",
       "2   -0.034807  0.131335 -0.169956  0.116559  0.063441 -0.047805  0.292008   \n",
       "3   -0.063434  0.197272 -0.178226  0.124807  0.080779 -0.057945  0.309352   \n",
       "4   -0.057145  0.168876 -0.176145  0.141866  0.069709 -0.062727  0.325756   \n",
       "..        ...       ...       ...       ...       ...       ...       ...   \n",
       "338  0.002627  0.223638 -0.272455  0.100111  0.011716 -0.065513  0.277845   \n",
       "339 -0.087806  0.161482 -0.118832  0.130527  0.026239 -0.087232  0.309075   \n",
       "340 -0.028243  0.128014 -0.191003  0.104212  0.113486 -0.089644  0.287929   \n",
       "341 -0.093090  0.199076 -0.139710  0.115822  0.061500 -0.093693  0.303419   \n",
       "342 -0.012566  0.126707 -0.309104  0.137018  0.024018 -0.095259  0.323033   \n",
       "\n",
       "           46        47  ...      5070      5071      5072      5073  \\\n",
       "0    0.060487  0.080803  ... -0.290288 -0.138646  0.337517 -0.036846   \n",
       "1    0.103770  0.101100  ... -0.303995 -0.181548  0.189195 -0.035498   \n",
       "2    0.127569  0.090374  ... -0.341078 -0.151316  0.166382 -0.083772   \n",
       "3    0.085288  0.077937  ... -0.238969 -0.153715  0.178666 -0.043892   \n",
       "4    0.127122  0.116105  ... -0.263345 -0.160914  0.203860 -0.063003   \n",
       "..        ...       ...  ...       ...       ...       ...       ...   \n",
       "338  0.094852  0.099362  ... -0.321741 -0.201256  0.263864 -0.045683   \n",
       "339  0.131613  0.099836  ... -0.255017 -0.156844  0.188265 -0.047636   \n",
       "340  0.059788  0.115429  ... -0.308235 -0.154275  0.266652 -0.090237   \n",
       "341  0.101090  0.110945  ... -0.247683 -0.131133  0.114285 -0.049539   \n",
       "342  0.036483  0.085557  ... -0.350772 -0.232596  0.292454 -0.092167   \n",
       "\n",
       "         5074      5075      5076      5077      5078      5079      5080  \\\n",
       "0   -0.145865  0.145739 -0.013954 -0.022534  0.266587  0.210630 -0.026933   \n",
       "1   -0.048187  0.136608 -0.038978  0.003040  0.180409  0.252026 -0.034072   \n",
       "2   -0.031800  0.173639 -0.039404 -0.026641  0.183440  0.283105 -0.058492   \n",
       "3   -0.123913  0.146099 -0.052013 -0.003793  0.247806  0.245565 -0.039263   \n",
       "4   -0.075318  0.142121  0.016088 -0.019374  0.254149  0.238047 -0.070921   \n",
       "..        ...       ...       ...       ...       ...       ...       ...   \n",
       "338 -0.053043  0.132980 -0.056922  0.008876  0.237592  0.226631 -0.030301   \n",
       "339 -0.010267  0.135643  0.006043 -0.046831  0.220550  0.258802 -0.059154   \n",
       "340 -0.035987  0.118418 -0.051667  0.000337  0.229528  0.222554 -0.052118   \n",
       "341 -0.046005  0.125978 -0.016459 -0.006963  0.209881  0.270406 -0.059499   \n",
       "342 -0.047533  0.123823 -0.004531  0.018452  0.234012  0.212497 -0.085228   \n",
       "\n",
       "         5081      5082      5083      5084      5085      5086      5087  \\\n",
       "0    0.026307  0.076589  0.187688  0.110229 -0.061802 -0.139272  0.225796   \n",
       "1    0.067309  0.028784  0.119834  0.231834 -0.008835 -0.155341  0.164639   \n",
       "2    0.033385  0.051223  0.157715  0.203213 -0.015537 -0.231012  0.176929   \n",
       "3    0.074070  0.054889  0.122570  0.174041 -0.029537 -0.161622  0.190393   \n",
       "4    0.075543  0.025655  0.128815  0.166325 -0.076467 -0.168410  0.199253   \n",
       "..        ...       ...       ...       ...       ...       ...       ...   \n",
       "338  0.053813  0.082987  0.189639  0.077907 -0.007688 -0.217726  0.268569   \n",
       "339  0.113412  0.003682  0.101039  0.176312 -0.023857 -0.145524  0.163611   \n",
       "340  0.005027  0.032014  0.162373  0.125081 -0.104945 -0.201631  0.208292   \n",
       "341  0.091360  0.007245  0.104477  0.172886 -0.041010 -0.134691  0.171602   \n",
       "342  0.007345  0.057953  0.194741  0.074880 -0.031264 -0.179551  0.225836   \n",
       "\n",
       "         5088      5089      5090      5091      5092      5093      5094  \\\n",
       "0   -0.113380 -0.024664 -0.230150 -0.086779 -0.018946 -0.150796  0.282690   \n",
       "1   -0.088200 -0.141192 -0.107459 -0.126025 -0.100561 -0.043595  0.288866   \n",
       "2   -0.108814 -0.151691 -0.155576 -0.138166 -0.069419 -0.081773  0.287196   \n",
       "3   -0.111680 -0.081653 -0.096576 -0.078513 -0.077252 -0.082461  0.261394   \n",
       "4   -0.106218 -0.079883 -0.145478 -0.124827 -0.123468 -0.094550  0.282291   \n",
       "..        ...       ...       ...       ...       ...       ...       ...   \n",
       "338 -0.098079  0.037021 -0.225534 -0.057410 -0.062185 -0.145098  0.322035   \n",
       "339 -0.092816 -0.087014 -0.064904 -0.182796 -0.146189 -0.044013  0.285041   \n",
       "340 -0.101363 -0.047418 -0.151285 -0.122030 -0.051473 -0.128324  0.273887   \n",
       "341 -0.101408 -0.083674 -0.061620 -0.128539 -0.090669 -0.067588  0.277691   \n",
       "342 -0.087978  0.025476 -0.189744 -0.065764 -0.042931 -0.169635  0.280548   \n",
       "\n",
       "         5095      5096      5097      5098      5099      5100      5101  \\\n",
       "0    0.062186  0.217737  0.001308 -0.076220  0.029293  0.134988  0.126751   \n",
       "1   -0.020607  0.087057 -0.002016 -0.042150  0.094406  0.091099  0.142107   \n",
       "2   -0.045575  0.097953 -0.033838 -0.036365  0.125570  0.049084  0.162091   \n",
       "3    0.016504  0.065125 -0.017172 -0.042037  0.077371  0.101969  0.171801   \n",
       "4    0.004593  0.082240 -0.030777 -0.038029  0.067017  0.078184  0.161283   \n",
       "..        ...       ...       ...       ...       ...       ...       ...   \n",
       "338  0.081622  0.172986 -0.022200 -0.015189  0.098192  0.149023  0.118560   \n",
       "339  0.000105  0.048911 -0.003084 -0.068871  0.079498  0.110505  0.126899   \n",
       "340  0.032952  0.174311 -0.035650 -0.026444  0.056570  0.084502  0.140714   \n",
       "341 -0.028381  0.046004 -0.016129 -0.058107  0.096573  0.124325  0.192702   \n",
       "342  0.072033  0.232753 -0.033540  0.023878  0.049674  0.110939  0.091594   \n",
       "\n",
       "         5102      5103      5104      5105      5106      5107      5108  \\\n",
       "0   -0.051001 -0.078870 -0.066645 -0.088527 -0.066581 -0.284267 -0.055163   \n",
       "1   -0.021993 -0.063503  0.044875  0.009611 -0.126196 -0.259387 -0.037057   \n",
       "2    0.003630 -0.094018 -0.003765  0.019716 -0.172657 -0.253969 -0.060248   \n",
       "3    0.012216 -0.091692 -0.028345  0.013726 -0.096583 -0.347500 -0.079666   \n",
       "4   -0.034093 -0.067665 -0.023683 -0.008662 -0.087251 -0.313730 -0.051508   \n",
       "..        ...       ...       ...       ...       ...       ...       ...   \n",
       "338 -0.052649 -0.121595 -0.139058 -0.076863  0.023087 -0.283470 -0.061043   \n",
       "339  0.010283 -0.083637  0.003246  0.031959 -0.092911 -0.287391 -0.017080   \n",
       "340 -0.054317 -0.064304 -0.016965 -0.037114 -0.067974 -0.293219 -0.059548   \n",
       "341  0.022009 -0.109851  0.015895  0.049376 -0.090216 -0.352958 -0.018720   \n",
       "342 -0.054995 -0.092958 -0.159015 -0.081230  0.003357 -0.259860 -0.099152   \n",
       "\n",
       "         5109      5110      5111      5112      5113      5114      5115  \\\n",
       "0    0.033041 -0.139801 -0.113517  0.121989 -0.190335  0.056447 -0.192906   \n",
       "1    0.041199 -0.139124 -0.158397  0.033567 -0.222891  0.147657 -0.210249   \n",
       "2    0.032126 -0.131952 -0.182381  0.012476 -0.216764  0.082182 -0.167000   \n",
       "3    0.017933 -0.133635 -0.175840  0.021177 -0.237654  0.146838 -0.213037   \n",
       "4    0.005071 -0.155179 -0.159542  0.033873 -0.211086  0.123042 -0.241100   \n",
       "..        ...       ...       ...       ...       ...       ...       ...   \n",
       "338 -0.001715 -0.124384 -0.141402  0.084083 -0.126614  0.090866 -0.164546   \n",
       "339  0.023965 -0.184101 -0.144901  0.060435 -0.246207  0.152064 -0.207770   \n",
       "340  0.028274 -0.170386 -0.156650  0.046467 -0.211848  0.081548 -0.178534   \n",
       "341  0.004421 -0.178036 -0.131529  0.062716 -0.249751  0.126609 -0.227497   \n",
       "342  0.050142 -0.154672 -0.116942  0.155337 -0.169825  0.113419 -0.165286   \n",
       "\n",
       "         5116      5117      5118      5119  \n",
       "0    0.056169 -0.037111 -0.264767  0.061206  \n",
       "1   -0.039008 -0.049254 -0.220128  0.026089  \n",
       "2    0.017850 -0.070759 -0.238455  0.011200  \n",
       "3    0.014747 -0.115953 -0.227792  0.040570  \n",
       "4    0.001314 -0.086826 -0.217587  0.040132  \n",
       "..        ...       ...       ...       ...  \n",
       "338  0.101935 -0.093842 -0.241469  0.064410  \n",
       "339 -0.011471 -0.131945 -0.231710  0.057206  \n",
       "340  0.055623 -0.080532 -0.263759  0.021452  \n",
       "341 -0.025272 -0.121043 -0.241836  0.042768  \n",
       "342  0.070795  0.003111 -0.265938 -0.002164  \n",
       "\n",
       "[343 rows x 5122 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load embeddings from a pickle file\n",
    "df = pd.read_pickle('data/embedding/esm2_15B_pg1_1mut.pkl')\n",
    "df"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Predicting active variants with AMP MBC <= 8"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "      <th>17</th>\n",
       "      <th>18</th>\n",
       "      <th>19</th>\n",
       "      <th>20</th>\n",
       "      <th>21</th>\n",
       "      <th>22</th>\n",
       "      <th>23</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "      <th>26</th>\n",
       "      <th>27</th>\n",
       "      <th>28</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "      <th>33</th>\n",
       "      <th>34</th>\n",
       "      <th>35</th>\n",
       "      <th>36</th>\n",
       "      <th>37</th>\n",
       "      <th>38</th>\n",
       "      <th>39</th>\n",
       "      <th>40</th>\n",
       "      <th>41</th>\n",
       "      <th>42</th>\n",
       "      <th>43</th>\n",
       "      <th>44</th>\n",
       "      <th>45</th>\n",
       "      <th>46</th>\n",
       "      <th>47</th>\n",
       "      <th>48</th>\n",
       "      <th>49</th>\n",
       "      <th>...</th>\n",
       "      <th>5070</th>\n",
       "      <th>5071</th>\n",
       "      <th>5072</th>\n",
       "      <th>5073</th>\n",
       "      <th>5074</th>\n",
       "      <th>5075</th>\n",
       "      <th>5076</th>\n",
       "      <th>5077</th>\n",
       "      <th>5078</th>\n",
       "      <th>5079</th>\n",
       "      <th>5080</th>\n",
       "      <th>5081</th>\n",
       "      <th>5082</th>\n",
       "      <th>5083</th>\n",
       "      <th>5084</th>\n",
       "      <th>5085</th>\n",
       "      <th>5086</th>\n",
       "      <th>5087</th>\n",
       "      <th>5088</th>\n",
       "      <th>5089</th>\n",
       "      <th>5090</th>\n",
       "      <th>5091</th>\n",
       "      <th>5092</th>\n",
       "      <th>5093</th>\n",
       "      <th>5094</th>\n",
       "      <th>5095</th>\n",
       "      <th>5096</th>\n",
       "      <th>5097</th>\n",
       "      <th>5098</th>\n",
       "      <th>5099</th>\n",
       "      <th>5100</th>\n",
       "      <th>5101</th>\n",
       "      <th>5102</th>\n",
       "      <th>5103</th>\n",
       "      <th>5104</th>\n",
       "      <th>5105</th>\n",
       "      <th>5106</th>\n",
       "      <th>5107</th>\n",
       "      <th>5108</th>\n",
       "      <th>5109</th>\n",
       "      <th>5110</th>\n",
       "      <th>5111</th>\n",
       "      <th>5112</th>\n",
       "      <th>5113</th>\n",
       "      <th>5114</th>\n",
       "      <th>5115</th>\n",
       "      <th>5116</th>\n",
       "      <th>5117</th>\n",
       "      <th>5118</th>\n",
       "      <th>5119</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.139956</td>\n",
       "      <td>-0.125581</td>\n",
       "      <td>-0.076405</td>\n",
       "      <td>0.146426</td>\n",
       "      <td>-0.280027</td>\n",
       "      <td>0.177829</td>\n",
       "      <td>-0.103134</td>\n",
       "      <td>0.157826</td>\n",
       "      <td>-0.095869</td>\n",
       "      <td>0.238305</td>\n",
       "      <td>-0.166609</td>\n",
       "      <td>-0.106000</td>\n",
       "      <td>0.197645</td>\n",
       "      <td>0.147172</td>\n",
       "      <td>0.172095</td>\n",
       "      <td>-0.173993</td>\n",
       "      <td>0.049640</td>\n",
       "      <td>-0.142380</td>\n",
       "      <td>0.071343</td>\n",
       "      <td>-0.033491</td>\n",
       "      <td>0.175970</td>\n",
       "      <td>0.044800</td>\n",
       "      <td>-0.230723</td>\n",
       "      <td>0.116288</td>\n",
       "      <td>-0.228306</td>\n",
       "      <td>-0.045376</td>\n",
       "      <td>0.081696</td>\n",
       "      <td>-0.078286</td>\n",
       "      <td>0.017577</td>\n",
       "      <td>0.161617</td>\n",
       "      <td>-0.144766</td>\n",
       "      <td>-0.046883</td>\n",
       "      <td>-0.329819</td>\n",
       "      <td>-0.064292</td>\n",
       "      <td>-0.049258</td>\n",
       "      <td>0.030661</td>\n",
       "      <td>0.182056</td>\n",
       "      <td>-0.066748</td>\n",
       "      <td>0.331273</td>\n",
       "      <td>-0.024626</td>\n",
       "      <td>0.153560</td>\n",
       "      <td>-0.341233</td>\n",
       "      <td>0.098469</td>\n",
       "      <td>0.041766</td>\n",
       "      <td>-0.098336</td>\n",
       "      <td>0.335028</td>\n",
       "      <td>0.060487</td>\n",
       "      <td>0.080803</td>\n",
       "      <td>-0.175508</td>\n",
       "      <td>0.016926</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.290288</td>\n",
       "      <td>-0.138646</td>\n",
       "      <td>0.337517</td>\n",
       "      <td>-0.036846</td>\n",
       "      <td>-0.145865</td>\n",
       "      <td>0.145739</td>\n",
       "      <td>-0.013954</td>\n",
       "      <td>-0.022534</td>\n",
       "      <td>0.266587</td>\n",
       "      <td>0.210630</td>\n",
       "      <td>-0.026933</td>\n",
       "      <td>0.026307</td>\n",
       "      <td>0.076589</td>\n",
       "      <td>0.187688</td>\n",
       "      <td>0.110229</td>\n",
       "      <td>-0.061802</td>\n",
       "      <td>-0.139272</td>\n",
       "      <td>0.225796</td>\n",
       "      <td>-0.113380</td>\n",
       "      <td>-0.024664</td>\n",
       "      <td>-0.230150</td>\n",
       "      <td>-0.086779</td>\n",
       "      <td>-0.018946</td>\n",
       "      <td>-0.150796</td>\n",
       "      <td>0.282690</td>\n",
       "      <td>0.062186</td>\n",
       "      <td>0.217737</td>\n",
       "      <td>0.001308</td>\n",
       "      <td>-0.076220</td>\n",
       "      <td>0.029293</td>\n",
       "      <td>0.134988</td>\n",
       "      <td>0.126751</td>\n",
       "      <td>-0.051001</td>\n",
       "      <td>-0.078870</td>\n",
       "      <td>-0.066645</td>\n",
       "      <td>-0.088527</td>\n",
       "      <td>-0.066581</td>\n",
       "      <td>-0.284267</td>\n",
       "      <td>-0.055163</td>\n",
       "      <td>0.033041</td>\n",
       "      <td>-0.139801</td>\n",
       "      <td>-0.113517</td>\n",
       "      <td>0.121989</td>\n",
       "      <td>-0.190335</td>\n",
       "      <td>0.056447</td>\n",
       "      <td>-0.192906</td>\n",
       "      <td>0.056169</td>\n",
       "      <td>-0.037111</td>\n",
       "      <td>-0.264767</td>\n",
       "      <td>0.061206</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.130926</td>\n",
       "      <td>0.000480</td>\n",
       "      <td>-0.139269</td>\n",
       "      <td>0.132204</td>\n",
       "      <td>-0.236016</td>\n",
       "      <td>0.171966</td>\n",
       "      <td>-0.085120</td>\n",
       "      <td>0.153231</td>\n",
       "      <td>-0.031941</td>\n",
       "      <td>0.153898</td>\n",
       "      <td>-0.142295</td>\n",
       "      <td>-0.114512</td>\n",
       "      <td>0.133251</td>\n",
       "      <td>0.141893</td>\n",
       "      <td>0.076887</td>\n",
       "      <td>-0.163610</td>\n",
       "      <td>0.135277</td>\n",
       "      <td>-0.086710</td>\n",
       "      <td>0.105006</td>\n",
       "      <td>0.039981</td>\n",
       "      <td>0.171417</td>\n",
       "      <td>-0.014597</td>\n",
       "      <td>-0.170309</td>\n",
       "      <td>0.056443</td>\n",
       "      <td>-0.123541</td>\n",
       "      <td>-0.021217</td>\n",
       "      <td>0.097796</td>\n",
       "      <td>-0.054952</td>\n",
       "      <td>0.091123</td>\n",
       "      <td>0.158605</td>\n",
       "      <td>-0.163539</td>\n",
       "      <td>-0.032665</td>\n",
       "      <td>-0.253657</td>\n",
       "      <td>-0.142816</td>\n",
       "      <td>-0.084620</td>\n",
       "      <td>0.004738</td>\n",
       "      <td>0.186130</td>\n",
       "      <td>-0.062229</td>\n",
       "      <td>0.138385</td>\n",
       "      <td>-0.051106</td>\n",
       "      <td>0.132185</td>\n",
       "      <td>-0.129907</td>\n",
       "      <td>0.131162</td>\n",
       "      <td>0.019662</td>\n",
       "      <td>-0.074399</td>\n",
       "      <td>0.299450</td>\n",
       "      <td>0.103770</td>\n",
       "      <td>0.101100</td>\n",
       "      <td>-0.090511</td>\n",
       "      <td>0.077786</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.303995</td>\n",
       "      <td>-0.181548</td>\n",
       "      <td>0.189195</td>\n",
       "      <td>-0.035498</td>\n",
       "      <td>-0.048187</td>\n",
       "      <td>0.136608</td>\n",
       "      <td>-0.038978</td>\n",
       "      <td>0.003040</td>\n",
       "      <td>0.180409</td>\n",
       "      <td>0.252026</td>\n",
       "      <td>-0.034072</td>\n",
       "      <td>0.067309</td>\n",
       "      <td>0.028784</td>\n",
       "      <td>0.119834</td>\n",
       "      <td>0.231834</td>\n",
       "      <td>-0.008835</td>\n",
       "      <td>-0.155341</td>\n",
       "      <td>0.164639</td>\n",
       "      <td>-0.088200</td>\n",
       "      <td>-0.141192</td>\n",
       "      <td>-0.107459</td>\n",
       "      <td>-0.126025</td>\n",
       "      <td>-0.100561</td>\n",
       "      <td>-0.043595</td>\n",
       "      <td>0.288866</td>\n",
       "      <td>-0.020607</td>\n",
       "      <td>0.087057</td>\n",
       "      <td>-0.002016</td>\n",
       "      <td>-0.042150</td>\n",
       "      <td>0.094406</td>\n",
       "      <td>0.091099</td>\n",
       "      <td>0.142107</td>\n",
       "      <td>-0.021993</td>\n",
       "      <td>-0.063503</td>\n",
       "      <td>0.044875</td>\n",
       "      <td>0.009611</td>\n",
       "      <td>-0.126196</td>\n",
       "      <td>-0.259387</td>\n",
       "      <td>-0.037057</td>\n",
       "      <td>0.041199</td>\n",
       "      <td>-0.139124</td>\n",
       "      <td>-0.158397</td>\n",
       "      <td>0.033567</td>\n",
       "      <td>-0.222891</td>\n",
       "      <td>0.147657</td>\n",
       "      <td>-0.210249</td>\n",
       "      <td>-0.039008</td>\n",
       "      <td>-0.049254</td>\n",
       "      <td>-0.220128</td>\n",
       "      <td>0.026089</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.122494</td>\n",
       "      <td>0.015151</td>\n",
       "      <td>-0.113578</td>\n",
       "      <td>0.147190</td>\n",
       "      <td>-0.248815</td>\n",
       "      <td>0.153636</td>\n",
       "      <td>-0.082545</td>\n",
       "      <td>0.143023</td>\n",
       "      <td>-0.032666</td>\n",
       "      <td>0.198368</td>\n",
       "      <td>-0.141825</td>\n",
       "      <td>-0.072487</td>\n",
       "      <td>0.164319</td>\n",
       "      <td>0.155215</td>\n",
       "      <td>0.000999</td>\n",
       "      <td>-0.142618</td>\n",
       "      <td>0.028556</td>\n",
       "      <td>-0.091350</td>\n",
       "      <td>0.169577</td>\n",
       "      <td>-0.013866</td>\n",
       "      <td>0.158021</td>\n",
       "      <td>-0.034980</td>\n",
       "      <td>-0.236284</td>\n",
       "      <td>0.121333</td>\n",
       "      <td>-0.169987</td>\n",
       "      <td>0.009360</td>\n",
       "      <td>0.132464</td>\n",
       "      <td>-0.091888</td>\n",
       "      <td>0.103464</td>\n",
       "      <td>0.148294</td>\n",
       "      <td>-0.164126</td>\n",
       "      <td>-0.009009</td>\n",
       "      <td>-0.292603</td>\n",
       "      <td>-0.130214</td>\n",
       "      <td>-0.137328</td>\n",
       "      <td>0.002338</td>\n",
       "      <td>0.152275</td>\n",
       "      <td>-0.041424</td>\n",
       "      <td>0.209629</td>\n",
       "      <td>-0.034807</td>\n",
       "      <td>0.131335</td>\n",
       "      <td>-0.169956</td>\n",
       "      <td>0.116559</td>\n",
       "      <td>0.063441</td>\n",
       "      <td>-0.047805</td>\n",
       "      <td>0.292008</td>\n",
       "      <td>0.127569</td>\n",
       "      <td>0.090374</td>\n",
       "      <td>-0.093007</td>\n",
       "      <td>-0.003598</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.341078</td>\n",
       "      <td>-0.151316</td>\n",
       "      <td>0.166382</td>\n",
       "      <td>-0.083772</td>\n",
       "      <td>-0.031800</td>\n",
       "      <td>0.173639</td>\n",
       "      <td>-0.039404</td>\n",
       "      <td>-0.026641</td>\n",
       "      <td>0.183440</td>\n",
       "      <td>0.283105</td>\n",
       "      <td>-0.058492</td>\n",
       "      <td>0.033385</td>\n",
       "      <td>0.051223</td>\n",
       "      <td>0.157715</td>\n",
       "      <td>0.203213</td>\n",
       "      <td>-0.015537</td>\n",
       "      <td>-0.231012</td>\n",
       "      <td>0.176929</td>\n",
       "      <td>-0.108814</td>\n",
       "      <td>-0.151691</td>\n",
       "      <td>-0.155576</td>\n",
       "      <td>-0.138166</td>\n",
       "      <td>-0.069419</td>\n",
       "      <td>-0.081773</td>\n",
       "      <td>0.287196</td>\n",
       "      <td>-0.045575</td>\n",
       "      <td>0.097953</td>\n",
       "      <td>-0.033838</td>\n",
       "      <td>-0.036365</td>\n",
       "      <td>0.125570</td>\n",
       "      <td>0.049084</td>\n",
       "      <td>0.162091</td>\n",
       "      <td>0.003630</td>\n",
       "      <td>-0.094018</td>\n",
       "      <td>-0.003765</td>\n",
       "      <td>0.019716</td>\n",
       "      <td>-0.172657</td>\n",
       "      <td>-0.253969</td>\n",
       "      <td>-0.060248</td>\n",
       "      <td>0.032126</td>\n",
       "      <td>-0.131952</td>\n",
       "      <td>-0.182381</td>\n",
       "      <td>0.012476</td>\n",
       "      <td>-0.216764</td>\n",
       "      <td>0.082182</td>\n",
       "      <td>-0.167000</td>\n",
       "      <td>0.017850</td>\n",
       "      <td>-0.070759</td>\n",
       "      <td>-0.238455</td>\n",
       "      <td>0.011200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.137794</td>\n",
       "      <td>-0.026297</td>\n",
       "      <td>-0.100598</td>\n",
       "      <td>0.132454</td>\n",
       "      <td>-0.240978</td>\n",
       "      <td>0.123862</td>\n",
       "      <td>-0.108370</td>\n",
       "      <td>0.163518</td>\n",
       "      <td>-0.074148</td>\n",
       "      <td>0.135701</td>\n",
       "      <td>-0.147309</td>\n",
       "      <td>-0.161241</td>\n",
       "      <td>0.135352</td>\n",
       "      <td>0.173438</td>\n",
       "      <td>0.115235</td>\n",
       "      <td>-0.128381</td>\n",
       "      <td>0.062604</td>\n",
       "      <td>-0.100330</td>\n",
       "      <td>0.103954</td>\n",
       "      <td>0.043673</td>\n",
       "      <td>0.147754</td>\n",
       "      <td>0.030620</td>\n",
       "      <td>-0.217643</td>\n",
       "      <td>0.078211</td>\n",
       "      <td>-0.180088</td>\n",
       "      <td>-0.034668</td>\n",
       "      <td>0.076345</td>\n",
       "      <td>-0.030924</td>\n",
       "      <td>0.068764</td>\n",
       "      <td>0.133319</td>\n",
       "      <td>-0.132511</td>\n",
       "      <td>-0.038674</td>\n",
       "      <td>-0.252596</td>\n",
       "      <td>-0.127496</td>\n",
       "      <td>-0.098959</td>\n",
       "      <td>0.027856</td>\n",
       "      <td>0.196983</td>\n",
       "      <td>-0.067005</td>\n",
       "      <td>0.181675</td>\n",
       "      <td>-0.063434</td>\n",
       "      <td>0.197272</td>\n",
       "      <td>-0.178226</td>\n",
       "      <td>0.124807</td>\n",
       "      <td>0.080779</td>\n",
       "      <td>-0.057945</td>\n",
       "      <td>0.309352</td>\n",
       "      <td>0.085288</td>\n",
       "      <td>0.077937</td>\n",
       "      <td>-0.118641</td>\n",
       "      <td>0.097144</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.238969</td>\n",
       "      <td>-0.153715</td>\n",
       "      <td>0.178666</td>\n",
       "      <td>-0.043892</td>\n",
       "      <td>-0.123913</td>\n",
       "      <td>0.146099</td>\n",
       "      <td>-0.052013</td>\n",
       "      <td>-0.003793</td>\n",
       "      <td>0.247806</td>\n",
       "      <td>0.245565</td>\n",
       "      <td>-0.039263</td>\n",
       "      <td>0.074070</td>\n",
       "      <td>0.054889</td>\n",
       "      <td>0.122570</td>\n",
       "      <td>0.174041</td>\n",
       "      <td>-0.029537</td>\n",
       "      <td>-0.161622</td>\n",
       "      <td>0.190393</td>\n",
       "      <td>-0.111680</td>\n",
       "      <td>-0.081653</td>\n",
       "      <td>-0.096576</td>\n",
       "      <td>-0.078513</td>\n",
       "      <td>-0.077252</td>\n",
       "      <td>-0.082461</td>\n",
       "      <td>0.261394</td>\n",
       "      <td>0.016504</td>\n",
       "      <td>0.065125</td>\n",
       "      <td>-0.017172</td>\n",
       "      <td>-0.042037</td>\n",
       "      <td>0.077371</td>\n",
       "      <td>0.101969</td>\n",
       "      <td>0.171801</td>\n",
       "      <td>0.012216</td>\n",
       "      <td>-0.091692</td>\n",
       "      <td>-0.028345</td>\n",
       "      <td>0.013726</td>\n",
       "      <td>-0.096583</td>\n",
       "      <td>-0.347500</td>\n",
       "      <td>-0.079666</td>\n",
       "      <td>0.017933</td>\n",
       "      <td>-0.133635</td>\n",
       "      <td>-0.175840</td>\n",
       "      <td>0.021177</td>\n",
       "      <td>-0.237654</td>\n",
       "      <td>0.146838</td>\n",
       "      <td>-0.213037</td>\n",
       "      <td>0.014747</td>\n",
       "      <td>-0.115953</td>\n",
       "      <td>-0.227792</td>\n",
       "      <td>0.040570</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.108533</td>\n",
       "      <td>-0.000612</td>\n",
       "      <td>-0.119274</td>\n",
       "      <td>0.135785</td>\n",
       "      <td>-0.260022</td>\n",
       "      <td>0.144817</td>\n",
       "      <td>-0.134771</td>\n",
       "      <td>0.222083</td>\n",
       "      <td>-0.071226</td>\n",
       "      <td>0.149381</td>\n",
       "      <td>-0.132229</td>\n",
       "      <td>-0.142084</td>\n",
       "      <td>0.202546</td>\n",
       "      <td>0.169355</td>\n",
       "      <td>0.116519</td>\n",
       "      <td>-0.167061</td>\n",
       "      <td>0.080885</td>\n",
       "      <td>-0.093418</td>\n",
       "      <td>0.138645</td>\n",
       "      <td>0.043515</td>\n",
       "      <td>0.134149</td>\n",
       "      <td>0.033941</td>\n",
       "      <td>-0.208214</td>\n",
       "      <td>0.081299</td>\n",
       "      <td>-0.160418</td>\n",
       "      <td>-0.035513</td>\n",
       "      <td>0.064047</td>\n",
       "      <td>-0.049382</td>\n",
       "      <td>0.048944</td>\n",
       "      <td>0.185777</td>\n",
       "      <td>-0.110163</td>\n",
       "      <td>-0.055136</td>\n",
       "      <td>-0.245780</td>\n",
       "      <td>-0.125597</td>\n",
       "      <td>-0.079749</td>\n",
       "      <td>0.033510</td>\n",
       "      <td>0.154440</td>\n",
       "      <td>-0.052611</td>\n",
       "      <td>0.233359</td>\n",
       "      <td>-0.057145</td>\n",
       "      <td>0.168876</td>\n",
       "      <td>-0.176145</td>\n",
       "      <td>0.141866</td>\n",
       "      <td>0.069709</td>\n",
       "      <td>-0.062727</td>\n",
       "      <td>0.325756</td>\n",
       "      <td>0.127122</td>\n",
       "      <td>0.116105</td>\n",
       "      <td>-0.109790</td>\n",
       "      <td>0.070471</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.263345</td>\n",
       "      <td>-0.160914</td>\n",
       "      <td>0.203860</td>\n",
       "      <td>-0.063003</td>\n",
       "      <td>-0.075318</td>\n",
       "      <td>0.142121</td>\n",
       "      <td>0.016088</td>\n",
       "      <td>-0.019374</td>\n",
       "      <td>0.254149</td>\n",
       "      <td>0.238047</td>\n",
       "      <td>-0.070921</td>\n",
       "      <td>0.075543</td>\n",
       "      <td>0.025655</td>\n",
       "      <td>0.128815</td>\n",
       "      <td>0.166325</td>\n",
       "      <td>-0.076467</td>\n",
       "      <td>-0.168410</td>\n",
       "      <td>0.199253</td>\n",
       "      <td>-0.106218</td>\n",
       "      <td>-0.079883</td>\n",
       "      <td>-0.145478</td>\n",
       "      <td>-0.124827</td>\n",
       "      <td>-0.123468</td>\n",
       "      <td>-0.094550</td>\n",
       "      <td>0.282291</td>\n",
       "      <td>0.004593</td>\n",
       "      <td>0.082240</td>\n",
       "      <td>-0.030777</td>\n",
       "      <td>-0.038029</td>\n",
       "      <td>0.067017</td>\n",
       "      <td>0.078184</td>\n",
       "      <td>0.161283</td>\n",
       "      <td>-0.034093</td>\n",
       "      <td>-0.067665</td>\n",
       "      <td>-0.023683</td>\n",
       "      <td>-0.008662</td>\n",
       "      <td>-0.087251</td>\n",
       "      <td>-0.313730</td>\n",
       "      <td>-0.051508</td>\n",
       "      <td>0.005071</td>\n",
       "      <td>-0.155179</td>\n",
       "      <td>-0.159542</td>\n",
       "      <td>0.033873</td>\n",
       "      <td>-0.211086</td>\n",
       "      <td>0.123042</td>\n",
       "      <td>-0.241100</td>\n",
       "      <td>0.001314</td>\n",
       "      <td>-0.086826</td>\n",
       "      <td>-0.217587</td>\n",
       "      <td>0.040132</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 5120 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       0         1         2         3         4         5         6     \\\n",
       "0  0.139956 -0.125581 -0.076405  0.146426 -0.280027  0.177829 -0.103134   \n",
       "1  0.130926  0.000480 -0.139269  0.132204 -0.236016  0.171966 -0.085120   \n",
       "2  0.122494  0.015151 -0.113578  0.147190 -0.248815  0.153636 -0.082545   \n",
       "3  0.137794 -0.026297 -0.100598  0.132454 -0.240978  0.123862 -0.108370   \n",
       "4  0.108533 -0.000612 -0.119274  0.135785 -0.260022  0.144817 -0.134771   \n",
       "\n",
       "       7         8         9         10        11        12        13    \\\n",
       "0  0.157826 -0.095869  0.238305 -0.166609 -0.106000  0.197645  0.147172   \n",
       "1  0.153231 -0.031941  0.153898 -0.142295 -0.114512  0.133251  0.141893   \n",
       "2  0.143023 -0.032666  0.198368 -0.141825 -0.072487  0.164319  0.155215   \n",
       "3  0.163518 -0.074148  0.135701 -0.147309 -0.161241  0.135352  0.173438   \n",
       "4  0.222083 -0.071226  0.149381 -0.132229 -0.142084  0.202546  0.169355   \n",
       "\n",
       "       14        15        16        17        18        19        20    \\\n",
       "0  0.172095 -0.173993  0.049640 -0.142380  0.071343 -0.033491  0.175970   \n",
       "1  0.076887 -0.163610  0.135277 -0.086710  0.105006  0.039981  0.171417   \n",
       "2  0.000999 -0.142618  0.028556 -0.091350  0.169577 -0.013866  0.158021   \n",
       "3  0.115235 -0.128381  0.062604 -0.100330  0.103954  0.043673  0.147754   \n",
       "4  0.116519 -0.167061  0.080885 -0.093418  0.138645  0.043515  0.134149   \n",
       "\n",
       "       21        22        23        24        25        26        27    \\\n",
       "0  0.044800 -0.230723  0.116288 -0.228306 -0.045376  0.081696 -0.078286   \n",
       "1 -0.014597 -0.170309  0.056443 -0.123541 -0.021217  0.097796 -0.054952   \n",
       "2 -0.034980 -0.236284  0.121333 -0.169987  0.009360  0.132464 -0.091888   \n",
       "3  0.030620 -0.217643  0.078211 -0.180088 -0.034668  0.076345 -0.030924   \n",
       "4  0.033941 -0.208214  0.081299 -0.160418 -0.035513  0.064047 -0.049382   \n",
       "\n",
       "       28        29        30        31        32        33        34    \\\n",
       "0  0.017577  0.161617 -0.144766 -0.046883 -0.329819 -0.064292 -0.049258   \n",
       "1  0.091123  0.158605 -0.163539 -0.032665 -0.253657 -0.142816 -0.084620   \n",
       "2  0.103464  0.148294 -0.164126 -0.009009 -0.292603 -0.130214 -0.137328   \n",
       "3  0.068764  0.133319 -0.132511 -0.038674 -0.252596 -0.127496 -0.098959   \n",
       "4  0.048944  0.185777 -0.110163 -0.055136 -0.245780 -0.125597 -0.079749   \n",
       "\n",
       "       35        36        37        38        39        40        41    \\\n",
       "0  0.030661  0.182056 -0.066748  0.331273 -0.024626  0.153560 -0.341233   \n",
       "1  0.004738  0.186130 -0.062229  0.138385 -0.051106  0.132185 -0.129907   \n",
       "2  0.002338  0.152275 -0.041424  0.209629 -0.034807  0.131335 -0.169956   \n",
       "3  0.027856  0.196983 -0.067005  0.181675 -0.063434  0.197272 -0.178226   \n",
       "4  0.033510  0.154440 -0.052611  0.233359 -0.057145  0.168876 -0.176145   \n",
       "\n",
       "       42        43        44        45        46        47        48    \\\n",
       "0  0.098469  0.041766 -0.098336  0.335028  0.060487  0.080803 -0.175508   \n",
       "1  0.131162  0.019662 -0.074399  0.299450  0.103770  0.101100 -0.090511   \n",
       "2  0.116559  0.063441 -0.047805  0.292008  0.127569  0.090374 -0.093007   \n",
       "3  0.124807  0.080779 -0.057945  0.309352  0.085288  0.077937 -0.118641   \n",
       "4  0.141866  0.069709 -0.062727  0.325756  0.127122  0.116105 -0.109790   \n",
       "\n",
       "       49    ...      5070      5071      5072      5073      5074      5075  \\\n",
       "0  0.016926  ... -0.290288 -0.138646  0.337517 -0.036846 -0.145865  0.145739   \n",
       "1  0.077786  ... -0.303995 -0.181548  0.189195 -0.035498 -0.048187  0.136608   \n",
       "2 -0.003598  ... -0.341078 -0.151316  0.166382 -0.083772 -0.031800  0.173639   \n",
       "3  0.097144  ... -0.238969 -0.153715  0.178666 -0.043892 -0.123913  0.146099   \n",
       "4  0.070471  ... -0.263345 -0.160914  0.203860 -0.063003 -0.075318  0.142121   \n",
       "\n",
       "       5076      5077      5078      5079      5080      5081      5082  \\\n",
       "0 -0.013954 -0.022534  0.266587  0.210630 -0.026933  0.026307  0.076589   \n",
       "1 -0.038978  0.003040  0.180409  0.252026 -0.034072  0.067309  0.028784   \n",
       "2 -0.039404 -0.026641  0.183440  0.283105 -0.058492  0.033385  0.051223   \n",
       "3 -0.052013 -0.003793  0.247806  0.245565 -0.039263  0.074070  0.054889   \n",
       "4  0.016088 -0.019374  0.254149  0.238047 -0.070921  0.075543  0.025655   \n",
       "\n",
       "       5083      5084      5085      5086      5087      5088      5089  \\\n",
       "0  0.187688  0.110229 -0.061802 -0.139272  0.225796 -0.113380 -0.024664   \n",
       "1  0.119834  0.231834 -0.008835 -0.155341  0.164639 -0.088200 -0.141192   \n",
       "2  0.157715  0.203213 -0.015537 -0.231012  0.176929 -0.108814 -0.151691   \n",
       "3  0.122570  0.174041 -0.029537 -0.161622  0.190393 -0.111680 -0.081653   \n",
       "4  0.128815  0.166325 -0.076467 -0.168410  0.199253 -0.106218 -0.079883   \n",
       "\n",
       "       5090      5091      5092      5093      5094      5095      5096  \\\n",
       "0 -0.230150 -0.086779 -0.018946 -0.150796  0.282690  0.062186  0.217737   \n",
       "1 -0.107459 -0.126025 -0.100561 -0.043595  0.288866 -0.020607  0.087057   \n",
       "2 -0.155576 -0.138166 -0.069419 -0.081773  0.287196 -0.045575  0.097953   \n",
       "3 -0.096576 -0.078513 -0.077252 -0.082461  0.261394  0.016504  0.065125   \n",
       "4 -0.145478 -0.124827 -0.123468 -0.094550  0.282291  0.004593  0.082240   \n",
       "\n",
       "       5097      5098      5099      5100      5101      5102      5103  \\\n",
       "0  0.001308 -0.076220  0.029293  0.134988  0.126751 -0.051001 -0.078870   \n",
       "1 -0.002016 -0.042150  0.094406  0.091099  0.142107 -0.021993 -0.063503   \n",
       "2 -0.033838 -0.036365  0.125570  0.049084  0.162091  0.003630 -0.094018   \n",
       "3 -0.017172 -0.042037  0.077371  0.101969  0.171801  0.012216 -0.091692   \n",
       "4 -0.030777 -0.038029  0.067017  0.078184  0.161283 -0.034093 -0.067665   \n",
       "\n",
       "       5104      5105      5106      5107      5108      5109      5110  \\\n",
       "0 -0.066645 -0.088527 -0.066581 -0.284267 -0.055163  0.033041 -0.139801   \n",
       "1  0.044875  0.009611 -0.126196 -0.259387 -0.037057  0.041199 -0.139124   \n",
       "2 -0.003765  0.019716 -0.172657 -0.253969 -0.060248  0.032126 -0.131952   \n",
       "3 -0.028345  0.013726 -0.096583 -0.347500 -0.079666  0.017933 -0.133635   \n",
       "4 -0.023683 -0.008662 -0.087251 -0.313730 -0.051508  0.005071 -0.155179   \n",
       "\n",
       "       5111      5112      5113      5114      5115      5116      5117  \\\n",
       "0 -0.113517  0.121989 -0.190335  0.056447 -0.192906  0.056169 -0.037111   \n",
       "1 -0.158397  0.033567 -0.222891  0.147657 -0.210249 -0.039008 -0.049254   \n",
       "2 -0.182381  0.012476 -0.216764  0.082182 -0.167000  0.017850 -0.070759   \n",
       "3 -0.175840  0.021177 -0.237654  0.146838 -0.213037  0.014747 -0.115953   \n",
       "4 -0.159542  0.033873 -0.211086  0.123042 -0.241100  0.001314 -0.086826   \n",
       "\n",
       "       5118      5119  \n",
       "0 -0.264767  0.061206  \n",
       "1 -0.220128  0.026089  \n",
       "2 -0.238455  0.011200  \n",
       "3 -0.227792  0.040570  \n",
       "4 -0.217587  0.040132  \n",
       "\n",
       "[5 rows x 5120 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features = df.iloc[:,2:]\n",
    "features.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>3631</th>\n",
       "      <th>4026</th>\n",
       "      <th>4066</th>\n",
       "      <th>1826</th>\n",
       "      <th>36</th>\n",
       "      <th>1767</th>\n",
       "      <th>2064</th>\n",
       "      <th>946</th>\n",
       "      <th>1461</th>\n",
       "      <th>971</th>\n",
       "      <th>3920</th>\n",
       "      <th>2103</th>\n",
       "      <th>3417</th>\n",
       "      <th>2234</th>\n",
       "      <th>303</th>\n",
       "      <th>1773</th>\n",
       "      <th>1576</th>\n",
       "      <th>3447</th>\n",
       "      <th>2265</th>\n",
       "      <th>4011</th>\n",
       "      <th>4962</th>\n",
       "      <th>4351</th>\n",
       "      <th>4892</th>\n",
       "      <th>3418</th>\n",
       "      <th>2930</th>\n",
       "      <th>2111</th>\n",
       "      <th>2231</th>\n",
       "      <th>1995</th>\n",
       "      <th>2124</th>\n",
       "      <th>4919</th>\n",
       "      <th>1761</th>\n",
       "      <th>3286</th>\n",
       "      <th>3041</th>\n",
       "      <th>3313</th>\n",
       "      <th>3750</th>\n",
       "      <th>1864</th>\n",
       "      <th>274</th>\n",
       "      <th>475</th>\n",
       "      <th>964</th>\n",
       "      <th>2107</th>\n",
       "      <th>3235</th>\n",
       "      <th>3645</th>\n",
       "      <th>3332</th>\n",
       "      <th>657</th>\n",
       "      <th>3140</th>\n",
       "      <th>4553</th>\n",
       "      <th>1020</th>\n",
       "      <th>4694</th>\n",
       "      <th>837</th>\n",
       "      <th>3563</th>\n",
       "      <th>...</th>\n",
       "      <th>620</th>\n",
       "      <th>1142</th>\n",
       "      <th>1271</th>\n",
       "      <th>156</th>\n",
       "      <th>4401</th>\n",
       "      <th>985</th>\n",
       "      <th>5017</th>\n",
       "      <th>1794</th>\n",
       "      <th>2020</th>\n",
       "      <th>4661</th>\n",
       "      <th>3886</th>\n",
       "      <th>550</th>\n",
       "      <th>5074</th>\n",
       "      <th>3182</th>\n",
       "      <th>14</th>\n",
       "      <th>1190</th>\n",
       "      <th>4061</th>\n",
       "      <th>4971</th>\n",
       "      <th>4602</th>\n",
       "      <th>5078</th>\n",
       "      <th>2114</th>\n",
       "      <th>568</th>\n",
       "      <th>3863</th>\n",
       "      <th>285</th>\n",
       "      <th>1724</th>\n",
       "      <th>4664</th>\n",
       "      <th>1658</th>\n",
       "      <th>3877</th>\n",
       "      <th>2505</th>\n",
       "      <th>5088</th>\n",
       "      <th>4194</th>\n",
       "      <th>248</th>\n",
       "      <th>4621</th>\n",
       "      <th>454</th>\n",
       "      <th>1739</th>\n",
       "      <th>591</th>\n",
       "      <th>2138</th>\n",
       "      <th>2856</th>\n",
       "      <th>2184</th>\n",
       "      <th>4501</th>\n",
       "      <th>4798</th>\n",
       "      <th>487</th>\n",
       "      <th>471</th>\n",
       "      <th>4208</th>\n",
       "      <th>5097</th>\n",
       "      <th>5040</th>\n",
       "      <th>2513</th>\n",
       "      <th>161</th>\n",
       "      <th>3928</th>\n",
       "      <th>298</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.295925</td>\n",
       "      <td>-0.242120</td>\n",
       "      <td>0.167013</td>\n",
       "      <td>0.063869</td>\n",
       "      <td>0.182056</td>\n",
       "      <td>-0.081493</td>\n",
       "      <td>0.073252</td>\n",
       "      <td>-0.021091</td>\n",
       "      <td>0.000907</td>\n",
       "      <td>-0.126511</td>\n",
       "      <td>-0.198422</td>\n",
       "      <td>0.157931</td>\n",
       "      <td>-0.035073</td>\n",
       "      <td>-0.185176</td>\n",
       "      <td>0.217157</td>\n",
       "      <td>0.281587</td>\n",
       "      <td>0.038734</td>\n",
       "      <td>0.177745</td>\n",
       "      <td>0.144090</td>\n",
       "      <td>-0.138290</td>\n",
       "      <td>-0.129770</td>\n",
       "      <td>-0.270526</td>\n",
       "      <td>-0.088028</td>\n",
       "      <td>0.210553</td>\n",
       "      <td>0.105362</td>\n",
       "      <td>-0.111635</td>\n",
       "      <td>-0.059913</td>\n",
       "      <td>-0.027770</td>\n",
       "      <td>-0.095960</td>\n",
       "      <td>0.056549</td>\n",
       "      <td>0.042887</td>\n",
       "      <td>0.269902</td>\n",
       "      <td>-0.017413</td>\n",
       "      <td>-0.003312</td>\n",
       "      <td>-0.096829</td>\n",
       "      <td>-0.076103</td>\n",
       "      <td>-0.003425</td>\n",
       "      <td>-0.039563</td>\n",
       "      <td>0.070706</td>\n",
       "      <td>-0.000164</td>\n",
       "      <td>-0.137018</td>\n",
       "      <td>-0.114083</td>\n",
       "      <td>-0.033518</td>\n",
       "      <td>0.102217</td>\n",
       "      <td>-0.006107</td>\n",
       "      <td>-0.072734</td>\n",
       "      <td>0.089712</td>\n",
       "      <td>-0.014140</td>\n",
       "      <td>0.013101</td>\n",
       "      <td>0.165262</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.030247</td>\n",
       "      <td>0.140166</td>\n",
       "      <td>0.294648</td>\n",
       "      <td>-0.273808</td>\n",
       "      <td>0.108743</td>\n",
       "      <td>-0.052881</td>\n",
       "      <td>0.136800</td>\n",
       "      <td>-0.009918</td>\n",
       "      <td>0.147446</td>\n",
       "      <td>-0.101078</td>\n",
       "      <td>0.009228</td>\n",
       "      <td>0.008433</td>\n",
       "      <td>-0.145865</td>\n",
       "      <td>0.134994</td>\n",
       "      <td>0.172095</td>\n",
       "      <td>-0.166064</td>\n",
       "      <td>0.024571</td>\n",
       "      <td>0.156423</td>\n",
       "      <td>0.081344</td>\n",
       "      <td>0.266587</td>\n",
       "      <td>-0.068114</td>\n",
       "      <td>-0.481086</td>\n",
       "      <td>0.016473</td>\n",
       "      <td>-0.027372</td>\n",
       "      <td>0.112814</td>\n",
       "      <td>-0.057424</td>\n",
       "      <td>0.098685</td>\n",
       "      <td>0.200954</td>\n",
       "      <td>-0.160305</td>\n",
       "      <td>-0.113380</td>\n",
       "      <td>-0.173411</td>\n",
       "      <td>0.260358</td>\n",
       "      <td>0.152904</td>\n",
       "      <td>-0.012075</td>\n",
       "      <td>-0.139789</td>\n",
       "      <td>0.010012</td>\n",
       "      <td>-0.096958</td>\n",
       "      <td>0.104122</td>\n",
       "      <td>0.172950</td>\n",
       "      <td>0.247138</td>\n",
       "      <td>0.102225</td>\n",
       "      <td>0.188541</td>\n",
       "      <td>0.027510</td>\n",
       "      <td>-0.098444</td>\n",
       "      <td>0.001308</td>\n",
       "      <td>-0.000299</td>\n",
       "      <td>-0.048910</td>\n",
       "      <td>0.056056</td>\n",
       "      <td>-0.109122</td>\n",
       "      <td>0.081819</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.250831</td>\n",
       "      <td>-0.210576</td>\n",
       "      <td>0.060968</td>\n",
       "      <td>-0.009510</td>\n",
       "      <td>0.186130</td>\n",
       "      <td>-0.105435</td>\n",
       "      <td>-0.021967</td>\n",
       "      <td>-0.013764</td>\n",
       "      <td>0.080433</td>\n",
       "      <td>-0.034848</td>\n",
       "      <td>-0.154932</td>\n",
       "      <td>0.124403</td>\n",
       "      <td>0.043585</td>\n",
       "      <td>-0.171345</td>\n",
       "      <td>0.167718</td>\n",
       "      <td>0.126912</td>\n",
       "      <td>0.022994</td>\n",
       "      <td>0.158764</td>\n",
       "      <td>0.082624</td>\n",
       "      <td>-0.189613</td>\n",
       "      <td>-0.049135</td>\n",
       "      <td>-0.139440</td>\n",
       "      <td>-0.136599</td>\n",
       "      <td>0.228204</td>\n",
       "      <td>0.105021</td>\n",
       "      <td>-0.071652</td>\n",
       "      <td>-0.126368</td>\n",
       "      <td>0.006084</td>\n",
       "      <td>-0.115036</td>\n",
       "      <td>0.134146</td>\n",
       "      <td>0.047768</td>\n",
       "      <td>0.257552</td>\n",
       "      <td>-0.086875</td>\n",
       "      <td>0.024079</td>\n",
       "      <td>-0.065470</td>\n",
       "      <td>-0.091035</td>\n",
       "      <td>-0.041541</td>\n",
       "      <td>0.004994</td>\n",
       "      <td>0.155991</td>\n",
       "      <td>0.097429</td>\n",
       "      <td>-0.084823</td>\n",
       "      <td>-0.060201</td>\n",
       "      <td>-0.002158</td>\n",
       "      <td>-0.031266</td>\n",
       "      <td>-0.193706</td>\n",
       "      <td>-0.022264</td>\n",
       "      <td>0.084303</td>\n",
       "      <td>-0.013811</td>\n",
       "      <td>0.057100</td>\n",
       "      <td>0.223617</td>\n",
       "      <td>...</td>\n",
       "      <td>0.020615</td>\n",
       "      <td>0.069761</td>\n",
       "      <td>0.250253</td>\n",
       "      <td>-0.236020</td>\n",
       "      <td>0.120356</td>\n",
       "      <td>-0.123138</td>\n",
       "      <td>0.066734</td>\n",
       "      <td>0.050427</td>\n",
       "      <td>0.079264</td>\n",
       "      <td>-0.059357</td>\n",
       "      <td>0.042818</td>\n",
       "      <td>-0.035389</td>\n",
       "      <td>-0.048187</td>\n",
       "      <td>0.101306</td>\n",
       "      <td>0.076887</td>\n",
       "      <td>-0.100897</td>\n",
       "      <td>0.023816</td>\n",
       "      <td>0.140845</td>\n",
       "      <td>0.036889</td>\n",
       "      <td>0.180409</td>\n",
       "      <td>-0.087783</td>\n",
       "      <td>-0.378558</td>\n",
       "      <td>-0.012693</td>\n",
       "      <td>-0.000289</td>\n",
       "      <td>0.179430</td>\n",
       "      <td>-0.123188</td>\n",
       "      <td>0.020760</td>\n",
       "      <td>0.165790</td>\n",
       "      <td>-0.167110</td>\n",
       "      <td>-0.088200</td>\n",
       "      <td>-0.143638</td>\n",
       "      <td>0.139897</td>\n",
       "      <td>0.119676</td>\n",
       "      <td>-0.002877</td>\n",
       "      <td>-0.075977</td>\n",
       "      <td>0.027919</td>\n",
       "      <td>-0.073405</td>\n",
       "      <td>0.051188</td>\n",
       "      <td>0.133208</td>\n",
       "      <td>0.186449</td>\n",
       "      <td>0.059309</td>\n",
       "      <td>0.155844</td>\n",
       "      <td>0.074690</td>\n",
       "      <td>-0.063735</td>\n",
       "      <td>-0.002016</td>\n",
       "      <td>0.037642</td>\n",
       "      <td>-0.017852</td>\n",
       "      <td>0.024680</td>\n",
       "      <td>-0.076351</td>\n",
       "      <td>0.055377</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.189448</td>\n",
       "      <td>-0.198901</td>\n",
       "      <td>0.126283</td>\n",
       "      <td>-0.032547</td>\n",
       "      <td>0.152275</td>\n",
       "      <td>-0.101898</td>\n",
       "      <td>-0.050272</td>\n",
       "      <td>0.027468</td>\n",
       "      <td>0.088600</td>\n",
       "      <td>-0.056368</td>\n",
       "      <td>-0.154721</td>\n",
       "      <td>0.107647</td>\n",
       "      <td>0.055131</td>\n",
       "      <td>-0.206999</td>\n",
       "      <td>0.115218</td>\n",
       "      <td>0.123672</td>\n",
       "      <td>0.046325</td>\n",
       "      <td>0.120904</td>\n",
       "      <td>0.081414</td>\n",
       "      <td>-0.239348</td>\n",
       "      <td>-0.047712</td>\n",
       "      <td>-0.170929</td>\n",
       "      <td>-0.078417</td>\n",
       "      <td>0.176191</td>\n",
       "      <td>0.084944</td>\n",
       "      <td>-0.073507</td>\n",
       "      <td>-0.110983</td>\n",
       "      <td>-0.007991</td>\n",
       "      <td>-0.136806</td>\n",
       "      <td>0.125949</td>\n",
       "      <td>0.090306</td>\n",
       "      <td>0.315210</td>\n",
       "      <td>-0.109947</td>\n",
       "      <td>0.050130</td>\n",
       "      <td>-0.060447</td>\n",
       "      <td>-0.055377</td>\n",
       "      <td>0.028978</td>\n",
       "      <td>-0.050322</td>\n",
       "      <td>0.125004</td>\n",
       "      <td>-0.004303</td>\n",
       "      <td>-0.127592</td>\n",
       "      <td>-0.061179</td>\n",
       "      <td>0.015894</td>\n",
       "      <td>0.006144</td>\n",
       "      <td>-0.169938</td>\n",
       "      <td>-0.052995</td>\n",
       "      <td>0.123099</td>\n",
       "      <td>0.023100</td>\n",
       "      <td>0.033719</td>\n",
       "      <td>0.220058</td>\n",
       "      <td>...</td>\n",
       "      <td>0.100700</td>\n",
       "      <td>0.095771</td>\n",
       "      <td>0.226416</td>\n",
       "      <td>-0.194992</td>\n",
       "      <td>0.113335</td>\n",
       "      <td>-0.016810</td>\n",
       "      <td>0.048021</td>\n",
       "      <td>0.058570</td>\n",
       "      <td>0.042933</td>\n",
       "      <td>-0.075682</td>\n",
       "      <td>0.023360</td>\n",
       "      <td>0.020446</td>\n",
       "      <td>-0.031800</td>\n",
       "      <td>0.156271</td>\n",
       "      <td>0.000999</td>\n",
       "      <td>-0.027998</td>\n",
       "      <td>0.056344</td>\n",
       "      <td>0.168250</td>\n",
       "      <td>-0.002072</td>\n",
       "      <td>0.183440</td>\n",
       "      <td>-0.069149</td>\n",
       "      <td>-0.321463</td>\n",
       "      <td>-0.083802</td>\n",
       "      <td>0.035832</td>\n",
       "      <td>0.208959</td>\n",
       "      <td>-0.142064</td>\n",
       "      <td>0.020557</td>\n",
       "      <td>0.168984</td>\n",
       "      <td>-0.176259</td>\n",
       "      <td>-0.108814</td>\n",
       "      <td>-0.076506</td>\n",
       "      <td>0.181103</td>\n",
       "      <td>0.107856</td>\n",
       "      <td>-0.013030</td>\n",
       "      <td>-0.087866</td>\n",
       "      <td>0.027987</td>\n",
       "      <td>-0.108985</td>\n",
       "      <td>0.013377</td>\n",
       "      <td>0.136345</td>\n",
       "      <td>0.154380</td>\n",
       "      <td>0.038610</td>\n",
       "      <td>0.134776</td>\n",
       "      <td>0.122032</td>\n",
       "      <td>-0.136953</td>\n",
       "      <td>-0.033838</td>\n",
       "      <td>0.044689</td>\n",
       "      <td>-0.004334</td>\n",
       "      <td>-0.041895</td>\n",
       "      <td>-0.088479</td>\n",
       "      <td>0.012718</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.276478</td>\n",
       "      <td>-0.187747</td>\n",
       "      <td>0.119757</td>\n",
       "      <td>-0.004543</td>\n",
       "      <td>0.196983</td>\n",
       "      <td>-0.080688</td>\n",
       "      <td>0.025091</td>\n",
       "      <td>0.028455</td>\n",
       "      <td>0.033829</td>\n",
       "      <td>-0.053059</td>\n",
       "      <td>-0.146947</td>\n",
       "      <td>0.130832</td>\n",
       "      <td>0.003023</td>\n",
       "      <td>-0.180767</td>\n",
       "      <td>0.143554</td>\n",
       "      <td>0.163057</td>\n",
       "      <td>0.021029</td>\n",
       "      <td>0.144590</td>\n",
       "      <td>0.072318</td>\n",
       "      <td>-0.183481</td>\n",
       "      <td>-0.100414</td>\n",
       "      <td>-0.204029</td>\n",
       "      <td>-0.104086</td>\n",
       "      <td>0.204415</td>\n",
       "      <td>0.118218</td>\n",
       "      <td>-0.075167</td>\n",
       "      <td>-0.131791</td>\n",
       "      <td>0.003893</td>\n",
       "      <td>-0.061813</td>\n",
       "      <td>0.067725</td>\n",
       "      <td>0.057835</td>\n",
       "      <td>0.288460</td>\n",
       "      <td>-0.086537</td>\n",
       "      <td>0.033689</td>\n",
       "      <td>-0.086501</td>\n",
       "      <td>-0.060304</td>\n",
       "      <td>-0.063066</td>\n",
       "      <td>0.007112</td>\n",
       "      <td>0.130788</td>\n",
       "      <td>0.066832</td>\n",
       "      <td>-0.086608</td>\n",
       "      <td>-0.108636</td>\n",
       "      <td>0.019519</td>\n",
       "      <td>0.006068</td>\n",
       "      <td>-0.147062</td>\n",
       "      <td>-0.059386</td>\n",
       "      <td>0.088136</td>\n",
       "      <td>-0.001418</td>\n",
       "      <td>0.048026</td>\n",
       "      <td>0.225348</td>\n",
       "      <td>...</td>\n",
       "      <td>0.042089</td>\n",
       "      <td>0.124159</td>\n",
       "      <td>0.235368</td>\n",
       "      <td>-0.203245</td>\n",
       "      <td>0.158414</td>\n",
       "      <td>-0.086953</td>\n",
       "      <td>0.067648</td>\n",
       "      <td>0.029067</td>\n",
       "      <td>0.110109</td>\n",
       "      <td>-0.102808</td>\n",
       "      <td>0.036870</td>\n",
       "      <td>-0.029042</td>\n",
       "      <td>-0.123913</td>\n",
       "      <td>0.126110</td>\n",
       "      <td>0.115235</td>\n",
       "      <td>-0.078098</td>\n",
       "      <td>0.043701</td>\n",
       "      <td>0.161957</td>\n",
       "      <td>-0.006261</td>\n",
       "      <td>0.247806</td>\n",
       "      <td>-0.103637</td>\n",
       "      <td>-0.378868</td>\n",
       "      <td>-0.025485</td>\n",
       "      <td>0.046330</td>\n",
       "      <td>0.175725</td>\n",
       "      <td>-0.079216</td>\n",
       "      <td>0.073986</td>\n",
       "      <td>0.202340</td>\n",
       "      <td>-0.180341</td>\n",
       "      <td>-0.111680</td>\n",
       "      <td>-0.118043</td>\n",
       "      <td>0.164800</td>\n",
       "      <td>0.171505</td>\n",
       "      <td>0.019980</td>\n",
       "      <td>-0.116467</td>\n",
       "      <td>-0.011438</td>\n",
       "      <td>-0.077033</td>\n",
       "      <td>0.072112</td>\n",
       "      <td>0.174782</td>\n",
       "      <td>0.179137</td>\n",
       "      <td>0.059401</td>\n",
       "      <td>0.153913</td>\n",
       "      <td>0.089088</td>\n",
       "      <td>-0.073277</td>\n",
       "      <td>-0.017172</td>\n",
       "      <td>0.028460</td>\n",
       "      <td>-0.024447</td>\n",
       "      <td>0.037157</td>\n",
       "      <td>-0.059642</td>\n",
       "      <td>0.056421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.279444</td>\n",
       "      <td>-0.218099</td>\n",
       "      <td>0.094855</td>\n",
       "      <td>-0.010427</td>\n",
       "      <td>0.154440</td>\n",
       "      <td>-0.075613</td>\n",
       "      <td>0.012266</td>\n",
       "      <td>0.039593</td>\n",
       "      <td>0.035353</td>\n",
       "      <td>-0.074463</td>\n",
       "      <td>-0.133198</td>\n",
       "      <td>0.131153</td>\n",
       "      <td>0.069691</td>\n",
       "      <td>-0.205767</td>\n",
       "      <td>0.142595</td>\n",
       "      <td>0.183809</td>\n",
       "      <td>0.073213</td>\n",
       "      <td>0.124418</td>\n",
       "      <td>0.110956</td>\n",
       "      <td>-0.182650</td>\n",
       "      <td>-0.079995</td>\n",
       "      <td>-0.204073</td>\n",
       "      <td>-0.103563</td>\n",
       "      <td>0.198114</td>\n",
       "      <td>0.093801</td>\n",
       "      <td>-0.045722</td>\n",
       "      <td>-0.123679</td>\n",
       "      <td>0.018068</td>\n",
       "      <td>-0.061187</td>\n",
       "      <td>0.093881</td>\n",
       "      <td>0.077717</td>\n",
       "      <td>0.315325</td>\n",
       "      <td>-0.099694</td>\n",
       "      <td>0.049513</td>\n",
       "      <td>-0.080993</td>\n",
       "      <td>-0.078453</td>\n",
       "      <td>-0.018478</td>\n",
       "      <td>-0.009026</td>\n",
       "      <td>0.107672</td>\n",
       "      <td>0.087145</td>\n",
       "      <td>-0.097533</td>\n",
       "      <td>-0.077727</td>\n",
       "      <td>0.006013</td>\n",
       "      <td>0.048215</td>\n",
       "      <td>-0.142196</td>\n",
       "      <td>-0.050412</td>\n",
       "      <td>0.113042</td>\n",
       "      <td>0.023663</td>\n",
       "      <td>0.017357</td>\n",
       "      <td>0.224006</td>\n",
       "      <td>...</td>\n",
       "      <td>0.034306</td>\n",
       "      <td>0.112098</td>\n",
       "      <td>0.249430</td>\n",
       "      <td>-0.215000</td>\n",
       "      <td>0.116892</td>\n",
       "      <td>-0.083270</td>\n",
       "      <td>0.068630</td>\n",
       "      <td>0.015449</td>\n",
       "      <td>0.101451</td>\n",
       "      <td>-0.085345</td>\n",
       "      <td>0.013961</td>\n",
       "      <td>-0.030098</td>\n",
       "      <td>-0.075318</td>\n",
       "      <td>0.115586</td>\n",
       "      <td>0.116519</td>\n",
       "      <td>-0.103215</td>\n",
       "      <td>0.015413</td>\n",
       "      <td>0.171177</td>\n",
       "      <td>-0.013271</td>\n",
       "      <td>0.254149</td>\n",
       "      <td>-0.071556</td>\n",
       "      <td>-0.376954</td>\n",
       "      <td>-0.020957</td>\n",
       "      <td>0.006332</td>\n",
       "      <td>0.160409</td>\n",
       "      <td>-0.093421</td>\n",
       "      <td>0.074304</td>\n",
       "      <td>0.204933</td>\n",
       "      <td>-0.180158</td>\n",
       "      <td>-0.106218</td>\n",
       "      <td>-0.151465</td>\n",
       "      <td>0.144450</td>\n",
       "      <td>0.104504</td>\n",
       "      <td>0.008414</td>\n",
       "      <td>-0.076745</td>\n",
       "      <td>-0.037999</td>\n",
       "      <td>-0.122411</td>\n",
       "      <td>0.106093</td>\n",
       "      <td>0.155268</td>\n",
       "      <td>0.182269</td>\n",
       "      <td>0.068032</td>\n",
       "      <td>0.157561</td>\n",
       "      <td>0.068757</td>\n",
       "      <td>-0.094179</td>\n",
       "      <td>-0.030777</td>\n",
       "      <td>0.052607</td>\n",
       "      <td>-0.027019</td>\n",
       "      <td>0.001521</td>\n",
       "      <td>-0.105075</td>\n",
       "      <td>0.067698</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 403 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       3631      4026      4066      1826      36        1767      2064  \\\n",
       "0  0.295925 -0.242120  0.167013  0.063869  0.182056 -0.081493  0.073252   \n",
       "1  0.250831 -0.210576  0.060968 -0.009510  0.186130 -0.105435 -0.021967   \n",
       "2  0.189448 -0.198901  0.126283 -0.032547  0.152275 -0.101898 -0.050272   \n",
       "3  0.276478 -0.187747  0.119757 -0.004543  0.196983 -0.080688  0.025091   \n",
       "4  0.279444 -0.218099  0.094855 -0.010427  0.154440 -0.075613  0.012266   \n",
       "\n",
       "       946       1461      971       3920      2103      3417      2234  \\\n",
       "0 -0.021091  0.000907 -0.126511 -0.198422  0.157931 -0.035073 -0.185176   \n",
       "1 -0.013764  0.080433 -0.034848 -0.154932  0.124403  0.043585 -0.171345   \n",
       "2  0.027468  0.088600 -0.056368 -0.154721  0.107647  0.055131 -0.206999   \n",
       "3  0.028455  0.033829 -0.053059 -0.146947  0.130832  0.003023 -0.180767   \n",
       "4  0.039593  0.035353 -0.074463 -0.133198  0.131153  0.069691 -0.205767   \n",
       "\n",
       "       303       1773      1576      3447      2265      4011      4962  \\\n",
       "0  0.217157  0.281587  0.038734  0.177745  0.144090 -0.138290 -0.129770   \n",
       "1  0.167718  0.126912  0.022994  0.158764  0.082624 -0.189613 -0.049135   \n",
       "2  0.115218  0.123672  0.046325  0.120904  0.081414 -0.239348 -0.047712   \n",
       "3  0.143554  0.163057  0.021029  0.144590  0.072318 -0.183481 -0.100414   \n",
       "4  0.142595  0.183809  0.073213  0.124418  0.110956 -0.182650 -0.079995   \n",
       "\n",
       "       4351      4892      3418      2930      2111      2231      1995  \\\n",
       "0 -0.270526 -0.088028  0.210553  0.105362 -0.111635 -0.059913 -0.027770   \n",
       "1 -0.139440 -0.136599  0.228204  0.105021 -0.071652 -0.126368  0.006084   \n",
       "2 -0.170929 -0.078417  0.176191  0.084944 -0.073507 -0.110983 -0.007991   \n",
       "3 -0.204029 -0.104086  0.204415  0.118218 -0.075167 -0.131791  0.003893   \n",
       "4 -0.204073 -0.103563  0.198114  0.093801 -0.045722 -0.123679  0.018068   \n",
       "\n",
       "       2124      4919      1761      3286      3041      3313      3750  \\\n",
       "0 -0.095960  0.056549  0.042887  0.269902 -0.017413 -0.003312 -0.096829   \n",
       "1 -0.115036  0.134146  0.047768  0.257552 -0.086875  0.024079 -0.065470   \n",
       "2 -0.136806  0.125949  0.090306  0.315210 -0.109947  0.050130 -0.060447   \n",
       "3 -0.061813  0.067725  0.057835  0.288460 -0.086537  0.033689 -0.086501   \n",
       "4 -0.061187  0.093881  0.077717  0.315325 -0.099694  0.049513 -0.080993   \n",
       "\n",
       "       1864      274       475       964       2107      3235      3645  \\\n",
       "0 -0.076103 -0.003425 -0.039563  0.070706 -0.000164 -0.137018 -0.114083   \n",
       "1 -0.091035 -0.041541  0.004994  0.155991  0.097429 -0.084823 -0.060201   \n",
       "2 -0.055377  0.028978 -0.050322  0.125004 -0.004303 -0.127592 -0.061179   \n",
       "3 -0.060304 -0.063066  0.007112  0.130788  0.066832 -0.086608 -0.108636   \n",
       "4 -0.078453 -0.018478 -0.009026  0.107672  0.087145 -0.097533 -0.077727   \n",
       "\n",
       "       3332      657       3140      4553      1020      4694      837   \\\n",
       "0 -0.033518  0.102217 -0.006107 -0.072734  0.089712 -0.014140  0.013101   \n",
       "1 -0.002158 -0.031266 -0.193706 -0.022264  0.084303 -0.013811  0.057100   \n",
       "2  0.015894  0.006144 -0.169938 -0.052995  0.123099  0.023100  0.033719   \n",
       "3  0.019519  0.006068 -0.147062 -0.059386  0.088136 -0.001418  0.048026   \n",
       "4  0.006013  0.048215 -0.142196 -0.050412  0.113042  0.023663  0.017357   \n",
       "\n",
       "       3563  ...      620       1142      1271      156       4401      985   \\\n",
       "0  0.165262  ... -0.030247  0.140166  0.294648 -0.273808  0.108743 -0.052881   \n",
       "1  0.223617  ...  0.020615  0.069761  0.250253 -0.236020  0.120356 -0.123138   \n",
       "2  0.220058  ...  0.100700  0.095771  0.226416 -0.194992  0.113335 -0.016810   \n",
       "3  0.225348  ...  0.042089  0.124159  0.235368 -0.203245  0.158414 -0.086953   \n",
       "4  0.224006  ...  0.034306  0.112098  0.249430 -0.215000  0.116892 -0.083270   \n",
       "\n",
       "       5017      1794      2020      4661      3886      550       5074  \\\n",
       "0  0.136800 -0.009918  0.147446 -0.101078  0.009228  0.008433 -0.145865   \n",
       "1  0.066734  0.050427  0.079264 -0.059357  0.042818 -0.035389 -0.048187   \n",
       "2  0.048021  0.058570  0.042933 -0.075682  0.023360  0.020446 -0.031800   \n",
       "3  0.067648  0.029067  0.110109 -0.102808  0.036870 -0.029042 -0.123913   \n",
       "4  0.068630  0.015449  0.101451 -0.085345  0.013961 -0.030098 -0.075318   \n",
       "\n",
       "       3182      14        1190      4061      4971      4602      5078  \\\n",
       "0  0.134994  0.172095 -0.166064  0.024571  0.156423  0.081344  0.266587   \n",
       "1  0.101306  0.076887 -0.100897  0.023816  0.140845  0.036889  0.180409   \n",
       "2  0.156271  0.000999 -0.027998  0.056344  0.168250 -0.002072  0.183440   \n",
       "3  0.126110  0.115235 -0.078098  0.043701  0.161957 -0.006261  0.247806   \n",
       "4  0.115586  0.116519 -0.103215  0.015413  0.171177 -0.013271  0.254149   \n",
       "\n",
       "       2114      568       3863      285       1724      4664      1658  \\\n",
       "0 -0.068114 -0.481086  0.016473 -0.027372  0.112814 -0.057424  0.098685   \n",
       "1 -0.087783 -0.378558 -0.012693 -0.000289  0.179430 -0.123188  0.020760   \n",
       "2 -0.069149 -0.321463 -0.083802  0.035832  0.208959 -0.142064  0.020557   \n",
       "3 -0.103637 -0.378868 -0.025485  0.046330  0.175725 -0.079216  0.073986   \n",
       "4 -0.071556 -0.376954 -0.020957  0.006332  0.160409 -0.093421  0.074304   \n",
       "\n",
       "       3877      2505      5088      4194      248       4621      454   \\\n",
       "0  0.200954 -0.160305 -0.113380 -0.173411  0.260358  0.152904 -0.012075   \n",
       "1  0.165790 -0.167110 -0.088200 -0.143638  0.139897  0.119676 -0.002877   \n",
       "2  0.168984 -0.176259 -0.108814 -0.076506  0.181103  0.107856 -0.013030   \n",
       "3  0.202340 -0.180341 -0.111680 -0.118043  0.164800  0.171505  0.019980   \n",
       "4  0.204933 -0.180158 -0.106218 -0.151465  0.144450  0.104504  0.008414   \n",
       "\n",
       "       1739      591       2138      2856      2184      4501      4798  \\\n",
       "0 -0.139789  0.010012 -0.096958  0.104122  0.172950  0.247138  0.102225   \n",
       "1 -0.075977  0.027919 -0.073405  0.051188  0.133208  0.186449  0.059309   \n",
       "2 -0.087866  0.027987 -0.108985  0.013377  0.136345  0.154380  0.038610   \n",
       "3 -0.116467 -0.011438 -0.077033  0.072112  0.174782  0.179137  0.059401   \n",
       "4 -0.076745 -0.037999 -0.122411  0.106093  0.155268  0.182269  0.068032   \n",
       "\n",
       "       487       471       4208      5097      5040      2513      161   \\\n",
       "0  0.188541  0.027510 -0.098444  0.001308 -0.000299 -0.048910  0.056056   \n",
       "1  0.155844  0.074690 -0.063735 -0.002016  0.037642 -0.017852  0.024680   \n",
       "2  0.134776  0.122032 -0.136953 -0.033838  0.044689 -0.004334 -0.041895   \n",
       "3  0.153913  0.089088 -0.073277 -0.017172  0.028460 -0.024447  0.037157   \n",
       "4  0.157561  0.068757 -0.094179 -0.030777  0.052607 -0.027019  0.001521   \n",
       "\n",
       "       3928      298   \n",
       "0 -0.109122  0.081819  \n",
       "1 -0.076351  0.055377  \n",
       "2 -0.088479  0.012718  \n",
       "3 -0.059642  0.056421  \n",
       "4 -0.105075  0.067698  \n",
       "\n",
       "[5 rows x 403 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features1 = features[corr_columns_mic]\n",
    "features1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "222 peptides with probability == 100% to be active\n"
     ]
    }
   ],
   "source": [
    "# creates a df to save probabilities\n",
    "res_GNB = df.iloc[:,0:1].copy()\n",
    "\n",
    "# calculates the probability of being active\n",
    "prob= model_GNB_mbc.predict_proba(features1)\n",
    "prob= [(x[1]*100).round(2) for x in prob]\n",
    "\n",
    "res_GNB['MBC_8_prob']=prob\n",
    "\n",
    "#number of peptides with probability == 100%\n",
    "print(f\"{res_GNB.query('MBC_8_prob == 100').shape[0]} peptides with probability == 100% to be active\")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Predicting selectivity score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>2156</th>\n",
       "      <th>2623</th>\n",
       "      <th>3006</th>\n",
       "      <th>2530</th>\n",
       "      <th>3896</th>\n",
       "      <th>4531</th>\n",
       "      <th>502</th>\n",
       "      <th>3033</th>\n",
       "      <th>4603</th>\n",
       "      <th>1964</th>\n",
       "      <th>4090</th>\n",
       "      <th>2251</th>\n",
       "      <th>3284</th>\n",
       "      <th>1935</th>\n",
       "      <th>2761</th>\n",
       "      <th>4514</th>\n",
       "      <th>2951</th>\n",
       "      <th>2983</th>\n",
       "      <th>472</th>\n",
       "      <th>4801</th>\n",
       "      <th>4027</th>\n",
       "      <th>620</th>\n",
       "      <th>3448</th>\n",
       "      <th>2985</th>\n",
       "      <th>122</th>\n",
       "      <th>3205</th>\n",
       "      <th>4467</th>\n",
       "      <th>2917</th>\n",
       "      <th>4617</th>\n",
       "      <th>2742</th>\n",
       "      <th>1757</th>\n",
       "      <th>2963</th>\n",
       "      <th>1537</th>\n",
       "      <th>1484</th>\n",
       "      <th>4118</th>\n",
       "      <th>85</th>\n",
       "      <th>1571</th>\n",
       "      <th>1645</th>\n",
       "      <th>1012</th>\n",
       "      <th>1658</th>\n",
       "      <th>365</th>\n",
       "      <th>487</th>\n",
       "      <th>2045</th>\n",
       "      <th>5016</th>\n",
       "      <th>4667</th>\n",
       "      <th>1194</th>\n",
       "      <th>4320</th>\n",
       "      <th>254</th>\n",
       "      <th>3182</th>\n",
       "      <th>1090</th>\n",
       "      <th>2290</th>\n",
       "      <th>332</th>\n",
       "      <th>646</th>\n",
       "      <th>3668</th>\n",
       "      <th>2307</th>\n",
       "      <th>4691</th>\n",
       "      <th>3427</th>\n",
       "      <th>5040</th>\n",
       "      <th>507</th>\n",
       "      <th>4385</th>\n",
       "      <th>4664</th>\n",
       "      <th>1603</th>\n",
       "      <th>2231</th>\n",
       "      <th>197</th>\n",
       "      <th>1994</th>\n",
       "      <th>2210</th>\n",
       "      <th>3863</th>\n",
       "      <th>2560</th>\n",
       "      <th>1739</th>\n",
       "      <th>4885</th>\n",
       "      <th>2990</th>\n",
       "      <th>3775</th>\n",
       "      <th>312</th>\n",
       "      <th>4468</th>\n",
       "      <th>4707</th>\n",
       "      <th>1876</th>\n",
       "      <th>3895</th>\n",
       "      <th>609</th>\n",
       "      <th>4672</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.316914</td>\n",
       "      <td>0.133888</td>\n",
       "      <td>0.165485</td>\n",
       "      <td>0.219287</td>\n",
       "      <td>0.005795</td>\n",
       "      <td>0.138278</td>\n",
       "      <td>0.190562</td>\n",
       "      <td>0.100969</td>\n",
       "      <td>-0.198953</td>\n",
       "      <td>0.030877</td>\n",
       "      <td>-0.295455</td>\n",
       "      <td>-0.130345</td>\n",
       "      <td>0.001749</td>\n",
       "      <td>-0.071333</td>\n",
       "      <td>-0.073157</td>\n",
       "      <td>-0.325801</td>\n",
       "      <td>-0.005251</td>\n",
       "      <td>-0.134178</td>\n",
       "      <td>-0.177722</td>\n",
       "      <td>-0.121746</td>\n",
       "      <td>0.083719</td>\n",
       "      <td>-0.030247</td>\n",
       "      <td>0.095069</td>\n",
       "      <td>0.204233</td>\n",
       "      <td>-0.129875</td>\n",
       "      <td>-0.098153</td>\n",
       "      <td>-0.009536</td>\n",
       "      <td>0.157045</td>\n",
       "      <td>0.115093</td>\n",
       "      <td>0.012345</td>\n",
       "      <td>0.143856</td>\n",
       "      <td>0.103582</td>\n",
       "      <td>-0.083215</td>\n",
       "      <td>-0.187637</td>\n",
       "      <td>-0.060929</td>\n",
       "      <td>-0.192945</td>\n",
       "      <td>0.011973</td>\n",
       "      <td>0.064210</td>\n",
       "      <td>0.044195</td>\n",
       "      <td>0.098685</td>\n",
       "      <td>-0.133377</td>\n",
       "      <td>0.188541</td>\n",
       "      <td>0.095624</td>\n",
       "      <td>-0.180989</td>\n",
       "      <td>0.152122</td>\n",
       "      <td>0.082203</td>\n",
       "      <td>-0.213870</td>\n",
       "      <td>-0.127766</td>\n",
       "      <td>0.134994</td>\n",
       "      <td>0.008414</td>\n",
       "      <td>0.079548</td>\n",
       "      <td>-0.127265</td>\n",
       "      <td>0.272669</td>\n",
       "      <td>0.279339</td>\n",
       "      <td>-0.171858</td>\n",
       "      <td>-0.055522</td>\n",
       "      <td>0.040830</td>\n",
       "      <td>-0.000299</td>\n",
       "      <td>-0.048106</td>\n",
       "      <td>0.000320</td>\n",
       "      <td>-0.057424</td>\n",
       "      <td>-0.258994</td>\n",
       "      <td>-0.059913</td>\n",
       "      <td>-0.022343</td>\n",
       "      <td>0.227568</td>\n",
       "      <td>-0.116951</td>\n",
       "      <td>0.016473</td>\n",
       "      <td>-0.235236</td>\n",
       "      <td>-0.139789</td>\n",
       "      <td>0.083098</td>\n",
       "      <td>0.211681</td>\n",
       "      <td>0.103510</td>\n",
       "      <td>-0.098713</td>\n",
       "      <td>-0.336973</td>\n",
       "      <td>0.083653</td>\n",
       "      <td>0.321159</td>\n",
       "      <td>0.128183</td>\n",
       "      <td>-0.057011</td>\n",
       "      <td>-0.043930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.247514</td>\n",
       "      <td>0.110006</td>\n",
       "      <td>0.025019</td>\n",
       "      <td>0.167552</td>\n",
       "      <td>0.040785</td>\n",
       "      <td>0.013030</td>\n",
       "      <td>0.083217</td>\n",
       "      <td>-0.013882</td>\n",
       "      <td>-0.028441</td>\n",
       "      <td>-0.039021</td>\n",
       "      <td>-0.169919</td>\n",
       "      <td>-0.041991</td>\n",
       "      <td>0.055421</td>\n",
       "      <td>-0.021226</td>\n",
       "      <td>-0.074402</td>\n",
       "      <td>-0.277303</td>\n",
       "      <td>0.068798</td>\n",
       "      <td>-0.137941</td>\n",
       "      <td>-0.106809</td>\n",
       "      <td>-0.091816</td>\n",
       "      <td>0.028207</td>\n",
       "      <td>0.020615</td>\n",
       "      <td>0.021804</td>\n",
       "      <td>0.136305</td>\n",
       "      <td>-0.017604</td>\n",
       "      <td>-0.041836</td>\n",
       "      <td>-0.001940</td>\n",
       "      <td>0.238943</td>\n",
       "      <td>0.034722</td>\n",
       "      <td>-0.011705</td>\n",
       "      <td>-0.028060</td>\n",
       "      <td>0.041976</td>\n",
       "      <td>-0.025361</td>\n",
       "      <td>-0.183244</td>\n",
       "      <td>-0.145703</td>\n",
       "      <td>-0.090543</td>\n",
       "      <td>0.075615</td>\n",
       "      <td>0.022303</td>\n",
       "      <td>0.105810</td>\n",
       "      <td>0.020760</td>\n",
       "      <td>-0.050371</td>\n",
       "      <td>0.155844</td>\n",
       "      <td>0.011238</td>\n",
       "      <td>-0.191518</td>\n",
       "      <td>0.241090</td>\n",
       "      <td>0.093544</td>\n",
       "      <td>-0.131049</td>\n",
       "      <td>-0.180463</td>\n",
       "      <td>0.101306</td>\n",
       "      <td>-0.066947</td>\n",
       "      <td>0.078916</td>\n",
       "      <td>-0.078862</td>\n",
       "      <td>0.195092</td>\n",
       "      <td>0.153149</td>\n",
       "      <td>-0.075150</td>\n",
       "      <td>-0.108124</td>\n",
       "      <td>-0.075250</td>\n",
       "      <td>0.037642</td>\n",
       "      <td>-0.056348</td>\n",
       "      <td>0.016244</td>\n",
       "      <td>-0.123188</td>\n",
       "      <td>-0.194733</td>\n",
       "      <td>-0.126368</td>\n",
       "      <td>-0.169353</td>\n",
       "      <td>0.317833</td>\n",
       "      <td>-0.010017</td>\n",
       "      <td>-0.012693</td>\n",
       "      <td>-0.233748</td>\n",
       "      <td>-0.075977</td>\n",
       "      <td>-0.003912</td>\n",
       "      <td>0.118771</td>\n",
       "      <td>0.145114</td>\n",
       "      <td>-0.020273</td>\n",
       "      <td>-0.216888</td>\n",
       "      <td>0.032632</td>\n",
       "      <td>0.214837</td>\n",
       "      <td>0.019695</td>\n",
       "      <td>-0.035800</td>\n",
       "      <td>0.033320</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.200300</td>\n",
       "      <td>0.049459</td>\n",
       "      <td>0.065939</td>\n",
       "      <td>0.167296</td>\n",
       "      <td>0.037434</td>\n",
       "      <td>0.031106</td>\n",
       "      <td>0.108506</td>\n",
       "      <td>0.014452</td>\n",
       "      <td>-0.083313</td>\n",
       "      <td>-0.094369</td>\n",
       "      <td>-0.175193</td>\n",
       "      <td>-0.051597</td>\n",
       "      <td>0.059414</td>\n",
       "      <td>0.030091</td>\n",
       "      <td>-0.023948</td>\n",
       "      <td>-0.243607</td>\n",
       "      <td>0.068406</td>\n",
       "      <td>-0.166962</td>\n",
       "      <td>-0.137208</td>\n",
       "      <td>-0.114895</td>\n",
       "      <td>0.012523</td>\n",
       "      <td>0.100700</td>\n",
       "      <td>0.026789</td>\n",
       "      <td>0.135875</td>\n",
       "      <td>-0.010096</td>\n",
       "      <td>-0.008321</td>\n",
       "      <td>0.056658</td>\n",
       "      <td>0.214022</td>\n",
       "      <td>0.033992</td>\n",
       "      <td>-0.090221</td>\n",
       "      <td>-0.019194</td>\n",
       "      <td>0.076375</td>\n",
       "      <td>-0.099021</td>\n",
       "      <td>-0.234931</td>\n",
       "      <td>-0.153534</td>\n",
       "      <td>-0.117524</td>\n",
       "      <td>0.075873</td>\n",
       "      <td>0.043760</td>\n",
       "      <td>0.102304</td>\n",
       "      <td>0.020557</td>\n",
       "      <td>-0.020794</td>\n",
       "      <td>0.134776</td>\n",
       "      <td>0.036274</td>\n",
       "      <td>-0.153236</td>\n",
       "      <td>0.219400</td>\n",
       "      <td>0.046362</td>\n",
       "      <td>-0.130498</td>\n",
       "      <td>-0.262258</td>\n",
       "      <td>0.156271</td>\n",
       "      <td>-0.091339</td>\n",
       "      <td>0.047683</td>\n",
       "      <td>-0.085990</td>\n",
       "      <td>0.226329</td>\n",
       "      <td>0.174312</td>\n",
       "      <td>-0.101392</td>\n",
       "      <td>-0.073100</td>\n",
       "      <td>-0.103785</td>\n",
       "      <td>0.044689</td>\n",
       "      <td>-0.072596</td>\n",
       "      <td>0.082250</td>\n",
       "      <td>-0.142064</td>\n",
       "      <td>-0.233575</td>\n",
       "      <td>-0.110983</td>\n",
       "      <td>-0.186726</td>\n",
       "      <td>0.329537</td>\n",
       "      <td>-0.027105</td>\n",
       "      <td>-0.083802</td>\n",
       "      <td>-0.224664</td>\n",
       "      <td>-0.087866</td>\n",
       "      <td>0.002511</td>\n",
       "      <td>0.061816</td>\n",
       "      <td>0.220941</td>\n",
       "      <td>-0.020943</td>\n",
       "      <td>-0.225501</td>\n",
       "      <td>0.033557</td>\n",
       "      <td>0.271957</td>\n",
       "      <td>-0.001894</td>\n",
       "      <td>0.057719</td>\n",
       "      <td>0.059173</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.262885</td>\n",
       "      <td>0.122060</td>\n",
       "      <td>0.049670</td>\n",
       "      <td>0.220331</td>\n",
       "      <td>0.021971</td>\n",
       "      <td>0.016429</td>\n",
       "      <td>0.134832</td>\n",
       "      <td>-0.021566</td>\n",
       "      <td>-0.087972</td>\n",
       "      <td>-0.039192</td>\n",
       "      <td>-0.241906</td>\n",
       "      <td>-0.040926</td>\n",
       "      <td>0.107756</td>\n",
       "      <td>-0.040865</td>\n",
       "      <td>-0.069702</td>\n",
       "      <td>-0.277580</td>\n",
       "      <td>0.035794</td>\n",
       "      <td>-0.103118</td>\n",
       "      <td>-0.140858</td>\n",
       "      <td>-0.133212</td>\n",
       "      <td>0.028996</td>\n",
       "      <td>0.042089</td>\n",
       "      <td>0.031861</td>\n",
       "      <td>0.159370</td>\n",
       "      <td>-0.059679</td>\n",
       "      <td>-0.034349</td>\n",
       "      <td>0.026619</td>\n",
       "      <td>0.204531</td>\n",
       "      <td>0.047528</td>\n",
       "      <td>0.036264</td>\n",
       "      <td>-0.017368</td>\n",
       "      <td>0.082461</td>\n",
       "      <td>-0.022948</td>\n",
       "      <td>-0.183201</td>\n",
       "      <td>-0.167473</td>\n",
       "      <td>-0.070006</td>\n",
       "      <td>0.026955</td>\n",
       "      <td>0.036872</td>\n",
       "      <td>0.083279</td>\n",
       "      <td>0.073986</td>\n",
       "      <td>-0.053381</td>\n",
       "      <td>0.153913</td>\n",
       "      <td>0.058955</td>\n",
       "      <td>-0.200015</td>\n",
       "      <td>0.215191</td>\n",
       "      <td>0.074695</td>\n",
       "      <td>-0.121165</td>\n",
       "      <td>-0.183919</td>\n",
       "      <td>0.126110</td>\n",
       "      <td>0.000376</td>\n",
       "      <td>0.051635</td>\n",
       "      <td>-0.105172</td>\n",
       "      <td>0.260710</td>\n",
       "      <td>0.160587</td>\n",
       "      <td>-0.144557</td>\n",
       "      <td>-0.126532</td>\n",
       "      <td>-0.094587</td>\n",
       "      <td>0.028460</td>\n",
       "      <td>-0.060010</td>\n",
       "      <td>0.009505</td>\n",
       "      <td>-0.079216</td>\n",
       "      <td>-0.214275</td>\n",
       "      <td>-0.131791</td>\n",
       "      <td>-0.144703</td>\n",
       "      <td>0.319411</td>\n",
       "      <td>-0.022645</td>\n",
       "      <td>-0.025485</td>\n",
       "      <td>-0.221291</td>\n",
       "      <td>-0.116467</td>\n",
       "      <td>0.011188</td>\n",
       "      <td>0.161175</td>\n",
       "      <td>0.108426</td>\n",
       "      <td>-0.033655</td>\n",
       "      <td>-0.212589</td>\n",
       "      <td>0.030716</td>\n",
       "      <td>0.221345</td>\n",
       "      <td>0.084448</td>\n",
       "      <td>0.005288</td>\n",
       "      <td>0.026848</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.230725</td>\n",
       "      <td>0.067553</td>\n",
       "      <td>0.058238</td>\n",
       "      <td>0.239368</td>\n",
       "      <td>0.034779</td>\n",
       "      <td>0.015927</td>\n",
       "      <td>0.111428</td>\n",
       "      <td>0.020500</td>\n",
       "      <td>-0.094747</td>\n",
       "      <td>-0.038990</td>\n",
       "      <td>-0.229986</td>\n",
       "      <td>-0.080027</td>\n",
       "      <td>0.106712</td>\n",
       "      <td>-0.041797</td>\n",
       "      <td>-0.058070</td>\n",
       "      <td>-0.284382</td>\n",
       "      <td>0.047071</td>\n",
       "      <td>-0.101410</td>\n",
       "      <td>-0.131763</td>\n",
       "      <td>-0.103706</td>\n",
       "      <td>0.047438</td>\n",
       "      <td>0.034306</td>\n",
       "      <td>0.036145</td>\n",
       "      <td>0.157087</td>\n",
       "      <td>-0.062504</td>\n",
       "      <td>0.003385</td>\n",
       "      <td>0.050751</td>\n",
       "      <td>0.178180</td>\n",
       "      <td>0.068373</td>\n",
       "      <td>0.010175</td>\n",
       "      <td>-0.006472</td>\n",
       "      <td>0.043826</td>\n",
       "      <td>-0.075219</td>\n",
       "      <td>-0.170017</td>\n",
       "      <td>-0.155436</td>\n",
       "      <td>-0.076816</td>\n",
       "      <td>0.051769</td>\n",
       "      <td>0.047510</td>\n",
       "      <td>0.083834</td>\n",
       "      <td>0.074304</td>\n",
       "      <td>-0.014229</td>\n",
       "      <td>0.157561</td>\n",
       "      <td>0.078213</td>\n",
       "      <td>-0.175569</td>\n",
       "      <td>0.219197</td>\n",
       "      <td>0.091785</td>\n",
       "      <td>-0.161756</td>\n",
       "      <td>-0.192189</td>\n",
       "      <td>0.115586</td>\n",
       "      <td>-0.049176</td>\n",
       "      <td>0.053376</td>\n",
       "      <td>-0.097292</td>\n",
       "      <td>0.225143</td>\n",
       "      <td>0.215206</td>\n",
       "      <td>-0.142996</td>\n",
       "      <td>-0.103003</td>\n",
       "      <td>-0.064907</td>\n",
       "      <td>0.052607</td>\n",
       "      <td>-0.044552</td>\n",
       "      <td>0.020002</td>\n",
       "      <td>-0.093421</td>\n",
       "      <td>-0.187675</td>\n",
       "      <td>-0.123679</td>\n",
       "      <td>-0.144751</td>\n",
       "      <td>0.309063</td>\n",
       "      <td>-0.060433</td>\n",
       "      <td>-0.020957</td>\n",
       "      <td>-0.267416</td>\n",
       "      <td>-0.076745</td>\n",
       "      <td>0.036661</td>\n",
       "      <td>0.099933</td>\n",
       "      <td>0.122974</td>\n",
       "      <td>-0.043910</td>\n",
       "      <td>-0.216034</td>\n",
       "      <td>0.035102</td>\n",
       "      <td>0.246464</td>\n",
       "      <td>0.067614</td>\n",
       "      <td>-0.011114</td>\n",
       "      <td>-0.009163</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       2156      2623      3006      2530      3896      4531       502  \\\n",
       "0  0.316914  0.133888  0.165485  0.219287  0.005795  0.138278  0.190562   \n",
       "1  0.247514  0.110006  0.025019  0.167552  0.040785  0.013030  0.083217   \n",
       "2  0.200300  0.049459  0.065939  0.167296  0.037434  0.031106  0.108506   \n",
       "3  0.262885  0.122060  0.049670  0.220331  0.021971  0.016429  0.134832   \n",
       "4  0.230725  0.067553  0.058238  0.239368  0.034779  0.015927  0.111428   \n",
       "\n",
       "       3033      4603      1964      4090      2251      3284      1935  \\\n",
       "0  0.100969 -0.198953  0.030877 -0.295455 -0.130345  0.001749 -0.071333   \n",
       "1 -0.013882 -0.028441 -0.039021 -0.169919 -0.041991  0.055421 -0.021226   \n",
       "2  0.014452 -0.083313 -0.094369 -0.175193 -0.051597  0.059414  0.030091   \n",
       "3 -0.021566 -0.087972 -0.039192 -0.241906 -0.040926  0.107756 -0.040865   \n",
       "4  0.020500 -0.094747 -0.038990 -0.229986 -0.080027  0.106712 -0.041797   \n",
       "\n",
       "       2761      4514      2951      2983       472      4801      4027  \\\n",
       "0 -0.073157 -0.325801 -0.005251 -0.134178 -0.177722 -0.121746  0.083719   \n",
       "1 -0.074402 -0.277303  0.068798 -0.137941 -0.106809 -0.091816  0.028207   \n",
       "2 -0.023948 -0.243607  0.068406 -0.166962 -0.137208 -0.114895  0.012523   \n",
       "3 -0.069702 -0.277580  0.035794 -0.103118 -0.140858 -0.133212  0.028996   \n",
       "4 -0.058070 -0.284382  0.047071 -0.101410 -0.131763 -0.103706  0.047438   \n",
       "\n",
       "        620      3448      2985       122      3205      4467      2917  \\\n",
       "0 -0.030247  0.095069  0.204233 -0.129875 -0.098153 -0.009536  0.157045   \n",
       "1  0.020615  0.021804  0.136305 -0.017604 -0.041836 -0.001940  0.238943   \n",
       "2  0.100700  0.026789  0.135875 -0.010096 -0.008321  0.056658  0.214022   \n",
       "3  0.042089  0.031861  0.159370 -0.059679 -0.034349  0.026619  0.204531   \n",
       "4  0.034306  0.036145  0.157087 -0.062504  0.003385  0.050751  0.178180   \n",
       "\n",
       "       4617      2742      1757      2963      1537      1484      4118  \\\n",
       "0  0.115093  0.012345  0.143856  0.103582 -0.083215 -0.187637 -0.060929   \n",
       "1  0.034722 -0.011705 -0.028060  0.041976 -0.025361 -0.183244 -0.145703   \n",
       "2  0.033992 -0.090221 -0.019194  0.076375 -0.099021 -0.234931 -0.153534   \n",
       "3  0.047528  0.036264 -0.017368  0.082461 -0.022948 -0.183201 -0.167473   \n",
       "4  0.068373  0.010175 -0.006472  0.043826 -0.075219 -0.170017 -0.155436   \n",
       "\n",
       "         85      1571      1645      1012      1658       365       487  \\\n",
       "0 -0.192945  0.011973  0.064210  0.044195  0.098685 -0.133377  0.188541   \n",
       "1 -0.090543  0.075615  0.022303  0.105810  0.020760 -0.050371  0.155844   \n",
       "2 -0.117524  0.075873  0.043760  0.102304  0.020557 -0.020794  0.134776   \n",
       "3 -0.070006  0.026955  0.036872  0.083279  0.073986 -0.053381  0.153913   \n",
       "4 -0.076816  0.051769  0.047510  0.083834  0.074304 -0.014229  0.157561   \n",
       "\n",
       "       2045      5016      4667      1194      4320       254      3182  \\\n",
       "0  0.095624 -0.180989  0.152122  0.082203 -0.213870 -0.127766  0.134994   \n",
       "1  0.011238 -0.191518  0.241090  0.093544 -0.131049 -0.180463  0.101306   \n",
       "2  0.036274 -0.153236  0.219400  0.046362 -0.130498 -0.262258  0.156271   \n",
       "3  0.058955 -0.200015  0.215191  0.074695 -0.121165 -0.183919  0.126110   \n",
       "4  0.078213 -0.175569  0.219197  0.091785 -0.161756 -0.192189  0.115586   \n",
       "\n",
       "       1090      2290       332       646      3668      2307      4691  \\\n",
       "0  0.008414  0.079548 -0.127265  0.272669  0.279339 -0.171858 -0.055522   \n",
       "1 -0.066947  0.078916 -0.078862  0.195092  0.153149 -0.075150 -0.108124   \n",
       "2 -0.091339  0.047683 -0.085990  0.226329  0.174312 -0.101392 -0.073100   \n",
       "3  0.000376  0.051635 -0.105172  0.260710  0.160587 -0.144557 -0.126532   \n",
       "4 -0.049176  0.053376 -0.097292  0.225143  0.215206 -0.142996 -0.103003   \n",
       "\n",
       "       3427      5040       507      4385      4664      1603      2231  \\\n",
       "0  0.040830 -0.000299 -0.048106  0.000320 -0.057424 -0.258994 -0.059913   \n",
       "1 -0.075250  0.037642 -0.056348  0.016244 -0.123188 -0.194733 -0.126368   \n",
       "2 -0.103785  0.044689 -0.072596  0.082250 -0.142064 -0.233575 -0.110983   \n",
       "3 -0.094587  0.028460 -0.060010  0.009505 -0.079216 -0.214275 -0.131791   \n",
       "4 -0.064907  0.052607 -0.044552  0.020002 -0.093421 -0.187675 -0.123679   \n",
       "\n",
       "        197      1994      2210      3863      2560      1739      4885  \\\n",
       "0 -0.022343  0.227568 -0.116951  0.016473 -0.235236 -0.139789  0.083098   \n",
       "1 -0.169353  0.317833 -0.010017 -0.012693 -0.233748 -0.075977 -0.003912   \n",
       "2 -0.186726  0.329537 -0.027105 -0.083802 -0.224664 -0.087866  0.002511   \n",
       "3 -0.144703  0.319411 -0.022645 -0.025485 -0.221291 -0.116467  0.011188   \n",
       "4 -0.144751  0.309063 -0.060433 -0.020957 -0.267416 -0.076745  0.036661   \n",
       "\n",
       "       2990      3775       312      4468      4707      1876      3895  \\\n",
       "0  0.211681  0.103510 -0.098713 -0.336973  0.083653  0.321159  0.128183   \n",
       "1  0.118771  0.145114 -0.020273 -0.216888  0.032632  0.214837  0.019695   \n",
       "2  0.061816  0.220941 -0.020943 -0.225501  0.033557  0.271957 -0.001894   \n",
       "3  0.161175  0.108426 -0.033655 -0.212589  0.030716  0.221345  0.084448   \n",
       "4  0.099933  0.122974 -0.043910 -0.216034  0.035102  0.246464  0.067614   \n",
       "\n",
       "        609      4672  \n",
       "0 -0.057011 -0.043930  \n",
       "1 -0.035800  0.033320  \n",
       "2  0.057719  0.059173  \n",
       "3  0.005288  0.026848  \n",
       "4 -0.011114 -0.009163  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features2 = features[corr_columns_sel]\n",
    "features2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 peptides with selectivity score < 0.5\n"
     ]
    }
   ],
   "source": [
    "# creates a df to selectivity score\n",
    "res_sel = df.iloc[:,0:1].copy()\n",
    "\n",
    "# calculates the the selectivity score\n",
    "prob= model_SVR_sel.predict(features2)\n",
    "res_sel['selectivity_pred']= prob.round(2) \n",
    "\n",
    "#number of peptides with selectivity score < 0.5\n",
    "print(f\"{res_sel.query('selectivity_pred < 0.5').shape[0]} peptides with selectivity score < 0.5\")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Prediction of Hemolysis % to protegrin-1 variants"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>4358</th>\n",
       "      <th>1901</th>\n",
       "      <th>4619</th>\n",
       "      <th>1777</th>\n",
       "      <th>2856</th>\n",
       "      <th>4623</th>\n",
       "      <th>3128</th>\n",
       "      <th>4145</th>\n",
       "      <th>4299</th>\n",
       "      <th>4484</th>\n",
       "      <th>4435</th>\n",
       "      <th>3520</th>\n",
       "      <th>1584</th>\n",
       "      <th>3285</th>\n",
       "      <th>1739</th>\n",
       "      <th>472</th>\n",
       "      <th>3404</th>\n",
       "      <th>1944</th>\n",
       "      <th>2714</th>\n",
       "      <th>3896</th>\n",
       "      <th>3482</th>\n",
       "      <th>4304</th>\n",
       "      <th>2964</th>\n",
       "      <th>48</th>\n",
       "      <th>4936</th>\n",
       "      <th>1971</th>\n",
       "      <th>4514</th>\n",
       "      <th>1518</th>\n",
       "      <th>3316</th>\n",
       "      <th>2655</th>\n",
       "      <th>3788</th>\n",
       "      <th>3006</th>\n",
       "      <th>2267</th>\n",
       "      <th>838</th>\n",
       "      <th>2923</th>\n",
       "      <th>3775</th>\n",
       "      <th>3712</th>\n",
       "      <th>2789</th>\n",
       "      <th>312</th>\n",
       "      <th>1155</th>\n",
       "      <th>2397</th>\n",
       "      <th>2917</th>\n",
       "      <th>3284</th>\n",
       "      <th>4617</th>\n",
       "      <th>4642</th>\n",
       "      <th>2233</th>\n",
       "      <th>3586</th>\n",
       "      <th>1370</th>\n",
       "      <th>4030</th>\n",
       "      <th>1417</th>\n",
       "      <th>...</th>\n",
       "      <th>2826</th>\n",
       "      <th>1916</th>\n",
       "      <th>3701</th>\n",
       "      <th>2492</th>\n",
       "      <th>3457</th>\n",
       "      <th>1571</th>\n",
       "      <th>912</th>\n",
       "      <th>4032</th>\n",
       "      <th>2990</th>\n",
       "      <th>2939</th>\n",
       "      <th>3466</th>\n",
       "      <th>3033</th>\n",
       "      <th>4385</th>\n",
       "      <th>2607</th>\n",
       "      <th>2067</th>\n",
       "      <th>809</th>\n",
       "      <th>3448</th>\n",
       "      <th>4027</th>\n",
       "      <th>3473</th>\n",
       "      <th>4926</th>\n",
       "      <th>2045</th>\n",
       "      <th>2093</th>\n",
       "      <th>1487</th>\n",
       "      <th>2772</th>\n",
       "      <th>783</th>\n",
       "      <th>2284</th>\n",
       "      <th>3942</th>\n",
       "      <th>4402</th>\n",
       "      <th>4686</th>\n",
       "      <th>2951</th>\n",
       "      <th>3399</th>\n",
       "      <th>4948</th>\n",
       "      <th>4040</th>\n",
       "      <th>1336</th>\n",
       "      <th>3891</th>\n",
       "      <th>4603</th>\n",
       "      <th>3718</th>\n",
       "      <th>933</th>\n",
       "      <th>2210</th>\n",
       "      <th>365</th>\n",
       "      <th>935</th>\n",
       "      <th>3427</th>\n",
       "      <th>646</th>\n",
       "      <th>1586</th>\n",
       "      <th>197</th>\n",
       "      <th>1757</th>\n",
       "      <th>609</th>\n",
       "      <th>4885</th>\n",
       "      <th>4468</th>\n",
       "      <th>1012</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.217128</td>\n",
       "      <td>-0.186775</td>\n",
       "      <td>-0.207822</td>\n",
       "      <td>-0.890151</td>\n",
       "      <td>0.104122</td>\n",
       "      <td>0.145655</td>\n",
       "      <td>0.249626</td>\n",
       "      <td>-0.152060</td>\n",
       "      <td>-0.049228</td>\n",
       "      <td>0.186836</td>\n",
       "      <td>-0.038690</td>\n",
       "      <td>-0.008260</td>\n",
       "      <td>0.041406</td>\n",
       "      <td>0.066750</td>\n",
       "      <td>-0.139789</td>\n",
       "      <td>-0.177722</td>\n",
       "      <td>-0.279815</td>\n",
       "      <td>-0.089450</td>\n",
       "      <td>-0.051162</td>\n",
       "      <td>0.005795</td>\n",
       "      <td>-0.093823</td>\n",
       "      <td>-0.209805</td>\n",
       "      <td>0.058046</td>\n",
       "      <td>-0.175508</td>\n",
       "      <td>0.076435</td>\n",
       "      <td>0.040915</td>\n",
       "      <td>-0.325801</td>\n",
       "      <td>0.028973</td>\n",
       "      <td>-0.210691</td>\n",
       "      <td>-0.387280</td>\n",
       "      <td>0.083730</td>\n",
       "      <td>0.165485</td>\n",
       "      <td>-0.204953</td>\n",
       "      <td>0.038528</td>\n",
       "      <td>0.044122</td>\n",
       "      <td>0.103510</td>\n",
       "      <td>0.117223</td>\n",
       "      <td>0.164936</td>\n",
       "      <td>-0.098713</td>\n",
       "      <td>-0.270531</td>\n",
       "      <td>-0.152582</td>\n",
       "      <td>0.157045</td>\n",
       "      <td>0.001749</td>\n",
       "      <td>0.115093</td>\n",
       "      <td>-0.043062</td>\n",
       "      <td>0.043278</td>\n",
       "      <td>0.003744</td>\n",
       "      <td>-0.098315</td>\n",
       "      <td>-0.153271</td>\n",
       "      <td>-0.289544</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.167059</td>\n",
       "      <td>-0.091951</td>\n",
       "      <td>-0.074190</td>\n",
       "      <td>0.076805</td>\n",
       "      <td>-0.109499</td>\n",
       "      <td>0.011973</td>\n",
       "      <td>-0.079815</td>\n",
       "      <td>0.178448</td>\n",
       "      <td>0.211681</td>\n",
       "      <td>-0.063050</td>\n",
       "      <td>-0.016370</td>\n",
       "      <td>0.100969</td>\n",
       "      <td>0.000320</td>\n",
       "      <td>-0.430121</td>\n",
       "      <td>-0.083275</td>\n",
       "      <td>0.170391</td>\n",
       "      <td>0.095069</td>\n",
       "      <td>0.083719</td>\n",
       "      <td>-0.070069</td>\n",
       "      <td>-0.089523</td>\n",
       "      <td>0.095624</td>\n",
       "      <td>-0.068355</td>\n",
       "      <td>0.100181</td>\n",
       "      <td>-0.015495</td>\n",
       "      <td>-0.057562</td>\n",
       "      <td>0.032309</td>\n",
       "      <td>-0.078187</td>\n",
       "      <td>-0.046740</td>\n",
       "      <td>0.048576</td>\n",
       "      <td>-0.005251</td>\n",
       "      <td>0.000218</td>\n",
       "      <td>-0.120265</td>\n",
       "      <td>-0.099890</td>\n",
       "      <td>0.089451</td>\n",
       "      <td>0.154824</td>\n",
       "      <td>-0.198953</td>\n",
       "      <td>-0.090117</td>\n",
       "      <td>-0.045577</td>\n",
       "      <td>-0.116951</td>\n",
       "      <td>-0.133377</td>\n",
       "      <td>-0.051211</td>\n",
       "      <td>0.040830</td>\n",
       "      <td>0.272669</td>\n",
       "      <td>0.184653</td>\n",
       "      <td>-0.022343</td>\n",
       "      <td>0.143856</td>\n",
       "      <td>-0.057011</td>\n",
       "      <td>0.083098</td>\n",
       "      <td>-0.336973</td>\n",
       "      <td>0.044195</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.107648</td>\n",
       "      <td>-0.124903</td>\n",
       "      <td>-0.154989</td>\n",
       "      <td>-0.663329</td>\n",
       "      <td>0.051188</td>\n",
       "      <td>0.224679</td>\n",
       "      <td>0.157115</td>\n",
       "      <td>-0.213559</td>\n",
       "      <td>-0.181494</td>\n",
       "      <td>0.032451</td>\n",
       "      <td>-0.010583</td>\n",
       "      <td>-0.113410</td>\n",
       "      <td>0.124761</td>\n",
       "      <td>-0.005504</td>\n",
       "      <td>-0.075977</td>\n",
       "      <td>-0.106809</td>\n",
       "      <td>-0.184245</td>\n",
       "      <td>-0.049736</td>\n",
       "      <td>0.120021</td>\n",
       "      <td>0.040785</td>\n",
       "      <td>0.050217</td>\n",
       "      <td>-0.265365</td>\n",
       "      <td>0.073642</td>\n",
       "      <td>-0.090511</td>\n",
       "      <td>-0.028179</td>\n",
       "      <td>-0.001556</td>\n",
       "      <td>-0.277303</td>\n",
       "      <td>-0.036342</td>\n",
       "      <td>-0.012253</td>\n",
       "      <td>-0.276300</td>\n",
       "      <td>-0.036408</td>\n",
       "      <td>0.025019</td>\n",
       "      <td>-0.297348</td>\n",
       "      <td>0.144452</td>\n",
       "      <td>-0.055525</td>\n",
       "      <td>0.145114</td>\n",
       "      <td>0.043523</td>\n",
       "      <td>0.274255</td>\n",
       "      <td>-0.020273</td>\n",
       "      <td>-0.174019</td>\n",
       "      <td>-0.027996</td>\n",
       "      <td>0.238943</td>\n",
       "      <td>0.055421</td>\n",
       "      <td>0.034722</td>\n",
       "      <td>0.050471</td>\n",
       "      <td>0.179608</td>\n",
       "      <td>0.062844</td>\n",
       "      <td>-0.041966</td>\n",
       "      <td>-0.066140</td>\n",
       "      <td>-0.269395</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.073628</td>\n",
       "      <td>0.011395</td>\n",
       "      <td>-0.202569</td>\n",
       "      <td>-0.095291</td>\n",
       "      <td>-0.183615</td>\n",
       "      <td>0.075615</td>\n",
       "      <td>-0.167485</td>\n",
       "      <td>0.042790</td>\n",
       "      <td>0.118771</td>\n",
       "      <td>0.155203</td>\n",
       "      <td>-0.110939</td>\n",
       "      <td>-0.013882</td>\n",
       "      <td>0.016244</td>\n",
       "      <td>-0.371703</td>\n",
       "      <td>0.013657</td>\n",
       "      <td>0.109096</td>\n",
       "      <td>0.021804</td>\n",
       "      <td>0.028207</td>\n",
       "      <td>0.004879</td>\n",
       "      <td>0.022190</td>\n",
       "      <td>0.011238</td>\n",
       "      <td>0.085485</td>\n",
       "      <td>-0.018277</td>\n",
       "      <td>-0.103544</td>\n",
       "      <td>0.056716</td>\n",
       "      <td>0.071136</td>\n",
       "      <td>-0.076833</td>\n",
       "      <td>-0.160793</td>\n",
       "      <td>0.172737</td>\n",
       "      <td>0.068798</td>\n",
       "      <td>0.126776</td>\n",
       "      <td>0.036691</td>\n",
       "      <td>-0.035327</td>\n",
       "      <td>0.143387</td>\n",
       "      <td>0.252215</td>\n",
       "      <td>-0.028441</td>\n",
       "      <td>-0.064453</td>\n",
       "      <td>-0.111541</td>\n",
       "      <td>-0.010017</td>\n",
       "      <td>-0.050371</td>\n",
       "      <td>-0.128424</td>\n",
       "      <td>-0.075250</td>\n",
       "      <td>0.195092</td>\n",
       "      <td>0.243829</td>\n",
       "      <td>-0.169353</td>\n",
       "      <td>-0.028060</td>\n",
       "      <td>-0.035800</td>\n",
       "      <td>-0.003912</td>\n",
       "      <td>-0.216888</td>\n",
       "      <td>0.105810</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.096685</td>\n",
       "      <td>-0.117793</td>\n",
       "      <td>-0.160030</td>\n",
       "      <td>-0.688115</td>\n",
       "      <td>0.013377</td>\n",
       "      <td>0.185849</td>\n",
       "      <td>0.203057</td>\n",
       "      <td>-0.243054</td>\n",
       "      <td>-0.131509</td>\n",
       "      <td>0.091401</td>\n",
       "      <td>-0.026314</td>\n",
       "      <td>-0.052209</td>\n",
       "      <td>0.131111</td>\n",
       "      <td>-0.006600</td>\n",
       "      <td>-0.087866</td>\n",
       "      <td>-0.137208</td>\n",
       "      <td>-0.232953</td>\n",
       "      <td>-0.010652</td>\n",
       "      <td>0.078191</td>\n",
       "      <td>0.037434</td>\n",
       "      <td>0.045446</td>\n",
       "      <td>-0.286628</td>\n",
       "      <td>0.053022</td>\n",
       "      <td>-0.093007</td>\n",
       "      <td>0.004643</td>\n",
       "      <td>-0.009288</td>\n",
       "      <td>-0.243607</td>\n",
       "      <td>-0.031015</td>\n",
       "      <td>0.006433</td>\n",
       "      <td>-0.268628</td>\n",
       "      <td>-0.045081</td>\n",
       "      <td>0.065939</td>\n",
       "      <td>-0.311183</td>\n",
       "      <td>0.136877</td>\n",
       "      <td>-0.073788</td>\n",
       "      <td>0.220941</td>\n",
       "      <td>0.021477</td>\n",
       "      <td>0.261384</td>\n",
       "      <td>-0.020943</td>\n",
       "      <td>-0.170651</td>\n",
       "      <td>-0.054040</td>\n",
       "      <td>0.214022</td>\n",
       "      <td>0.059414</td>\n",
       "      <td>0.033992</td>\n",
       "      <td>0.039593</td>\n",
       "      <td>0.108461</td>\n",
       "      <td>0.056378</td>\n",
       "      <td>-0.056969</td>\n",
       "      <td>-0.057147</td>\n",
       "      <td>-0.230433</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.071105</td>\n",
       "      <td>-0.003303</td>\n",
       "      <td>-0.184348</td>\n",
       "      <td>-0.103320</td>\n",
       "      <td>-0.203652</td>\n",
       "      <td>0.075873</td>\n",
       "      <td>-0.171245</td>\n",
       "      <td>0.059183</td>\n",
       "      <td>0.061816</td>\n",
       "      <td>0.183560</td>\n",
       "      <td>-0.105249</td>\n",
       "      <td>0.014452</td>\n",
       "      <td>0.082250</td>\n",
       "      <td>-0.340766</td>\n",
       "      <td>0.005703</td>\n",
       "      <td>0.092241</td>\n",
       "      <td>0.026789</td>\n",
       "      <td>0.012523</td>\n",
       "      <td>-0.005141</td>\n",
       "      <td>-0.052626</td>\n",
       "      <td>0.036274</td>\n",
       "      <td>0.137320</td>\n",
       "      <td>0.008819</td>\n",
       "      <td>-0.153893</td>\n",
       "      <td>0.080561</td>\n",
       "      <td>0.079028</td>\n",
       "      <td>-0.059329</td>\n",
       "      <td>-0.200058</td>\n",
       "      <td>0.158811</td>\n",
       "      <td>0.068406</td>\n",
       "      <td>0.134488</td>\n",
       "      <td>0.066705</td>\n",
       "      <td>-0.034635</td>\n",
       "      <td>0.116200</td>\n",
       "      <td>0.271434</td>\n",
       "      <td>-0.083313</td>\n",
       "      <td>-0.035127</td>\n",
       "      <td>-0.120600</td>\n",
       "      <td>-0.027105</td>\n",
       "      <td>-0.020794</td>\n",
       "      <td>-0.122072</td>\n",
       "      <td>-0.103785</td>\n",
       "      <td>0.226329</td>\n",
       "      <td>0.246068</td>\n",
       "      <td>-0.186726</td>\n",
       "      <td>-0.019194</td>\n",
       "      <td>0.057719</td>\n",
       "      <td>0.002511</td>\n",
       "      <td>-0.225501</td>\n",
       "      <td>0.102304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.113175</td>\n",
       "      <td>-0.154276</td>\n",
       "      <td>-0.132477</td>\n",
       "      <td>-0.652215</td>\n",
       "      <td>0.072112</td>\n",
       "      <td>0.203771</td>\n",
       "      <td>0.208632</td>\n",
       "      <td>-0.232611</td>\n",
       "      <td>-0.141950</td>\n",
       "      <td>0.077290</td>\n",
       "      <td>-0.018136</td>\n",
       "      <td>-0.109010</td>\n",
       "      <td>0.105711</td>\n",
       "      <td>0.005725</td>\n",
       "      <td>-0.116467</td>\n",
       "      <td>-0.140858</td>\n",
       "      <td>-0.201781</td>\n",
       "      <td>-0.027220</td>\n",
       "      <td>0.060352</td>\n",
       "      <td>0.021971</td>\n",
       "      <td>0.056460</td>\n",
       "      <td>-0.285506</td>\n",
       "      <td>0.067296</td>\n",
       "      <td>-0.118641</td>\n",
       "      <td>0.025931</td>\n",
       "      <td>0.046731</td>\n",
       "      <td>-0.277580</td>\n",
       "      <td>0.002781</td>\n",
       "      <td>-0.022896</td>\n",
       "      <td>-0.318044</td>\n",
       "      <td>-0.032710</td>\n",
       "      <td>0.049670</td>\n",
       "      <td>-0.268506</td>\n",
       "      <td>0.115405</td>\n",
       "      <td>-0.018237</td>\n",
       "      <td>0.108426</td>\n",
       "      <td>0.078003</td>\n",
       "      <td>0.235092</td>\n",
       "      <td>-0.033655</td>\n",
       "      <td>-0.204743</td>\n",
       "      <td>-0.084482</td>\n",
       "      <td>0.204531</td>\n",
       "      <td>0.107756</td>\n",
       "      <td>0.047528</td>\n",
       "      <td>0.041737</td>\n",
       "      <td>0.084542</td>\n",
       "      <td>0.066851</td>\n",
       "      <td>-0.102109</td>\n",
       "      <td>-0.059750</td>\n",
       "      <td>-0.312874</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.073863</td>\n",
       "      <td>0.060228</td>\n",
       "      <td>-0.132326</td>\n",
       "      <td>-0.002207</td>\n",
       "      <td>-0.185606</td>\n",
       "      <td>0.026955</td>\n",
       "      <td>-0.127501</td>\n",
       "      <td>0.078568</td>\n",
       "      <td>0.161175</td>\n",
       "      <td>0.112114</td>\n",
       "      <td>-0.117374</td>\n",
       "      <td>-0.021566</td>\n",
       "      <td>0.009505</td>\n",
       "      <td>-0.388407</td>\n",
       "      <td>0.008589</td>\n",
       "      <td>0.122895</td>\n",
       "      <td>0.031861</td>\n",
       "      <td>0.028996</td>\n",
       "      <td>-0.014724</td>\n",
       "      <td>-0.038751</td>\n",
       "      <td>0.058955</td>\n",
       "      <td>0.062241</td>\n",
       "      <td>0.029432</td>\n",
       "      <td>-0.115214</td>\n",
       "      <td>0.046334</td>\n",
       "      <td>0.080287</td>\n",
       "      <td>-0.093853</td>\n",
       "      <td>-0.151231</td>\n",
       "      <td>0.124380</td>\n",
       "      <td>0.035794</td>\n",
       "      <td>0.077359</td>\n",
       "      <td>0.003219</td>\n",
       "      <td>-0.041315</td>\n",
       "      <td>0.120044</td>\n",
       "      <td>0.172953</td>\n",
       "      <td>-0.087972</td>\n",
       "      <td>-0.077713</td>\n",
       "      <td>-0.099067</td>\n",
       "      <td>-0.022645</td>\n",
       "      <td>-0.053381</td>\n",
       "      <td>-0.102642</td>\n",
       "      <td>-0.094587</td>\n",
       "      <td>0.260710</td>\n",
       "      <td>0.183705</td>\n",
       "      <td>-0.144703</td>\n",
       "      <td>-0.017368</td>\n",
       "      <td>0.005288</td>\n",
       "      <td>0.011188</td>\n",
       "      <td>-0.212589</td>\n",
       "      <td>0.083279</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.105817</td>\n",
       "      <td>-0.156925</td>\n",
       "      <td>-0.134647</td>\n",
       "      <td>-0.696202</td>\n",
       "      <td>0.106093</td>\n",
       "      <td>0.164509</td>\n",
       "      <td>0.175014</td>\n",
       "      <td>-0.204543</td>\n",
       "      <td>-0.130316</td>\n",
       "      <td>0.087704</td>\n",
       "      <td>-0.001184</td>\n",
       "      <td>-0.098570</td>\n",
       "      <td>0.103609</td>\n",
       "      <td>-0.023367</td>\n",
       "      <td>-0.076745</td>\n",
       "      <td>-0.131763</td>\n",
       "      <td>-0.181690</td>\n",
       "      <td>-0.023427</td>\n",
       "      <td>0.045204</td>\n",
       "      <td>0.034779</td>\n",
       "      <td>0.034093</td>\n",
       "      <td>-0.300907</td>\n",
       "      <td>0.044663</td>\n",
       "      <td>-0.109790</td>\n",
       "      <td>0.016538</td>\n",
       "      <td>-0.004624</td>\n",
       "      <td>-0.284382</td>\n",
       "      <td>-0.011988</td>\n",
       "      <td>-0.035669</td>\n",
       "      <td>-0.294235</td>\n",
       "      <td>-0.020498</td>\n",
       "      <td>0.058238</td>\n",
       "      <td>-0.298005</td>\n",
       "      <td>0.088155</td>\n",
       "      <td>-0.075448</td>\n",
       "      <td>0.122974</td>\n",
       "      <td>0.082824</td>\n",
       "      <td>0.238809</td>\n",
       "      <td>-0.043910</td>\n",
       "      <td>-0.215907</td>\n",
       "      <td>-0.047443</td>\n",
       "      <td>0.178180</td>\n",
       "      <td>0.106712</td>\n",
       "      <td>0.068373</td>\n",
       "      <td>0.041550</td>\n",
       "      <td>0.088670</td>\n",
       "      <td>0.070121</td>\n",
       "      <td>-0.082190</td>\n",
       "      <td>-0.071344</td>\n",
       "      <td>-0.287204</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.108644</td>\n",
       "      <td>0.036899</td>\n",
       "      <td>-0.099915</td>\n",
       "      <td>-0.031518</td>\n",
       "      <td>-0.167010</td>\n",
       "      <td>0.051769</td>\n",
       "      <td>-0.108590</td>\n",
       "      <td>0.054865</td>\n",
       "      <td>0.099933</td>\n",
       "      <td>0.077882</td>\n",
       "      <td>-0.082707</td>\n",
       "      <td>0.020500</td>\n",
       "      <td>0.020002</td>\n",
       "      <td>-0.385074</td>\n",
       "      <td>-0.015226</td>\n",
       "      <td>0.096307</td>\n",
       "      <td>0.036145</td>\n",
       "      <td>0.047438</td>\n",
       "      <td>-0.011217</td>\n",
       "      <td>-0.044019</td>\n",
       "      <td>0.078213</td>\n",
       "      <td>0.071704</td>\n",
       "      <td>0.021166</td>\n",
       "      <td>-0.084150</td>\n",
       "      <td>0.027801</td>\n",
       "      <td>0.069000</td>\n",
       "      <td>-0.105269</td>\n",
       "      <td>-0.163250</td>\n",
       "      <td>0.131664</td>\n",
       "      <td>0.047071</td>\n",
       "      <td>0.084197</td>\n",
       "      <td>-0.012405</td>\n",
       "      <td>-0.041364</td>\n",
       "      <td>0.104473</td>\n",
       "      <td>0.175623</td>\n",
       "      <td>-0.094747</td>\n",
       "      <td>-0.056366</td>\n",
       "      <td>-0.099124</td>\n",
       "      <td>-0.060433</td>\n",
       "      <td>-0.014229</td>\n",
       "      <td>-0.094919</td>\n",
       "      <td>-0.064907</td>\n",
       "      <td>0.225143</td>\n",
       "      <td>0.203515</td>\n",
       "      <td>-0.144751</td>\n",
       "      <td>-0.006472</td>\n",
       "      <td>-0.011114</td>\n",
       "      <td>0.036661</td>\n",
       "      <td>-0.216034</td>\n",
       "      <td>0.083834</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 115 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       4358      1901      4619      1777      2856      4623      3128  \\\n",
       "0  0.217128 -0.186775 -0.207822 -0.890151  0.104122  0.145655  0.249626   \n",
       "1  0.107648 -0.124903 -0.154989 -0.663329  0.051188  0.224679  0.157115   \n",
       "2  0.096685 -0.117793 -0.160030 -0.688115  0.013377  0.185849  0.203057   \n",
       "3  0.113175 -0.154276 -0.132477 -0.652215  0.072112  0.203771  0.208632   \n",
       "4  0.105817 -0.156925 -0.134647 -0.696202  0.106093  0.164509  0.175014   \n",
       "\n",
       "       4145      4299      4484      4435      3520      1584      3285  \\\n",
       "0 -0.152060 -0.049228  0.186836 -0.038690 -0.008260  0.041406  0.066750   \n",
       "1 -0.213559 -0.181494  0.032451 -0.010583 -0.113410  0.124761 -0.005504   \n",
       "2 -0.243054 -0.131509  0.091401 -0.026314 -0.052209  0.131111 -0.006600   \n",
       "3 -0.232611 -0.141950  0.077290 -0.018136 -0.109010  0.105711  0.005725   \n",
       "4 -0.204543 -0.130316  0.087704 -0.001184 -0.098570  0.103609 -0.023367   \n",
       "\n",
       "       1739      472       3404      1944      2714      3896      3482  \\\n",
       "0 -0.139789 -0.177722 -0.279815 -0.089450 -0.051162  0.005795 -0.093823   \n",
       "1 -0.075977 -0.106809 -0.184245 -0.049736  0.120021  0.040785  0.050217   \n",
       "2 -0.087866 -0.137208 -0.232953 -0.010652  0.078191  0.037434  0.045446   \n",
       "3 -0.116467 -0.140858 -0.201781 -0.027220  0.060352  0.021971  0.056460   \n",
       "4 -0.076745 -0.131763 -0.181690 -0.023427  0.045204  0.034779  0.034093   \n",
       "\n",
       "       4304      2964      48        4936      1971      4514      1518  \\\n",
       "0 -0.209805  0.058046 -0.175508  0.076435  0.040915 -0.325801  0.028973   \n",
       "1 -0.265365  0.073642 -0.090511 -0.028179 -0.001556 -0.277303 -0.036342   \n",
       "2 -0.286628  0.053022 -0.093007  0.004643 -0.009288 -0.243607 -0.031015   \n",
       "3 -0.285506  0.067296 -0.118641  0.025931  0.046731 -0.277580  0.002781   \n",
       "4 -0.300907  0.044663 -0.109790  0.016538 -0.004624 -0.284382 -0.011988   \n",
       "\n",
       "       3316      2655      3788      3006      2267      838       2923  \\\n",
       "0 -0.210691 -0.387280  0.083730  0.165485 -0.204953  0.038528  0.044122   \n",
       "1 -0.012253 -0.276300 -0.036408  0.025019 -0.297348  0.144452 -0.055525   \n",
       "2  0.006433 -0.268628 -0.045081  0.065939 -0.311183  0.136877 -0.073788   \n",
       "3 -0.022896 -0.318044 -0.032710  0.049670 -0.268506  0.115405 -0.018237   \n",
       "4 -0.035669 -0.294235 -0.020498  0.058238 -0.298005  0.088155 -0.075448   \n",
       "\n",
       "       3775      3712      2789      312       1155      2397      2917  \\\n",
       "0  0.103510  0.117223  0.164936 -0.098713 -0.270531 -0.152582  0.157045   \n",
       "1  0.145114  0.043523  0.274255 -0.020273 -0.174019 -0.027996  0.238943   \n",
       "2  0.220941  0.021477  0.261384 -0.020943 -0.170651 -0.054040  0.214022   \n",
       "3  0.108426  0.078003  0.235092 -0.033655 -0.204743 -0.084482  0.204531   \n",
       "4  0.122974  0.082824  0.238809 -0.043910 -0.215907 -0.047443  0.178180   \n",
       "\n",
       "       3284      4617      4642      2233      3586      1370      4030  \\\n",
       "0  0.001749  0.115093 -0.043062  0.043278  0.003744 -0.098315 -0.153271   \n",
       "1  0.055421  0.034722  0.050471  0.179608  0.062844 -0.041966 -0.066140   \n",
       "2  0.059414  0.033992  0.039593  0.108461  0.056378 -0.056969 -0.057147   \n",
       "3  0.107756  0.047528  0.041737  0.084542  0.066851 -0.102109 -0.059750   \n",
       "4  0.106712  0.068373  0.041550  0.088670  0.070121 -0.082190 -0.071344   \n",
       "\n",
       "       1417  ...      2826      1916      3701      2492      3457      1571  \\\n",
       "0 -0.289544  ... -0.167059 -0.091951 -0.074190  0.076805 -0.109499  0.011973   \n",
       "1 -0.269395  ... -0.073628  0.011395 -0.202569 -0.095291 -0.183615  0.075615   \n",
       "2 -0.230433  ... -0.071105 -0.003303 -0.184348 -0.103320 -0.203652  0.075873   \n",
       "3 -0.312874  ... -0.073863  0.060228 -0.132326 -0.002207 -0.185606  0.026955   \n",
       "4 -0.287204  ... -0.108644  0.036899 -0.099915 -0.031518 -0.167010  0.051769   \n",
       "\n",
       "       912       4032      2990      2939      3466      3033      4385  \\\n",
       "0 -0.079815  0.178448  0.211681 -0.063050 -0.016370  0.100969  0.000320   \n",
       "1 -0.167485  0.042790  0.118771  0.155203 -0.110939 -0.013882  0.016244   \n",
       "2 -0.171245  0.059183  0.061816  0.183560 -0.105249  0.014452  0.082250   \n",
       "3 -0.127501  0.078568  0.161175  0.112114 -0.117374 -0.021566  0.009505   \n",
       "4 -0.108590  0.054865  0.099933  0.077882 -0.082707  0.020500  0.020002   \n",
       "\n",
       "       2607      2067      809       3448      4027      3473      4926  \\\n",
       "0 -0.430121 -0.083275  0.170391  0.095069  0.083719 -0.070069 -0.089523   \n",
       "1 -0.371703  0.013657  0.109096  0.021804  0.028207  0.004879  0.022190   \n",
       "2 -0.340766  0.005703  0.092241  0.026789  0.012523 -0.005141 -0.052626   \n",
       "3 -0.388407  0.008589  0.122895  0.031861  0.028996 -0.014724 -0.038751   \n",
       "4 -0.385074 -0.015226  0.096307  0.036145  0.047438 -0.011217 -0.044019   \n",
       "\n",
       "       2045      2093      1487      2772      783       2284      3942  \\\n",
       "0  0.095624 -0.068355  0.100181 -0.015495 -0.057562  0.032309 -0.078187   \n",
       "1  0.011238  0.085485 -0.018277 -0.103544  0.056716  0.071136 -0.076833   \n",
       "2  0.036274  0.137320  0.008819 -0.153893  0.080561  0.079028 -0.059329   \n",
       "3  0.058955  0.062241  0.029432 -0.115214  0.046334  0.080287 -0.093853   \n",
       "4  0.078213  0.071704  0.021166 -0.084150  0.027801  0.069000 -0.105269   \n",
       "\n",
       "       4402      4686      2951      3399      4948      4040      1336  \\\n",
       "0 -0.046740  0.048576 -0.005251  0.000218 -0.120265 -0.099890  0.089451   \n",
       "1 -0.160793  0.172737  0.068798  0.126776  0.036691 -0.035327  0.143387   \n",
       "2 -0.200058  0.158811  0.068406  0.134488  0.066705 -0.034635  0.116200   \n",
       "3 -0.151231  0.124380  0.035794  0.077359  0.003219 -0.041315  0.120044   \n",
       "4 -0.163250  0.131664  0.047071  0.084197 -0.012405 -0.041364  0.104473   \n",
       "\n",
       "       3891      4603      3718      933       2210      365       935   \\\n",
       "0  0.154824 -0.198953 -0.090117 -0.045577 -0.116951 -0.133377 -0.051211   \n",
       "1  0.252215 -0.028441 -0.064453 -0.111541 -0.010017 -0.050371 -0.128424   \n",
       "2  0.271434 -0.083313 -0.035127 -0.120600 -0.027105 -0.020794 -0.122072   \n",
       "3  0.172953 -0.087972 -0.077713 -0.099067 -0.022645 -0.053381 -0.102642   \n",
       "4  0.175623 -0.094747 -0.056366 -0.099124 -0.060433 -0.014229 -0.094919   \n",
       "\n",
       "       3427      646       1586      197       1757      609       4885  \\\n",
       "0  0.040830  0.272669  0.184653 -0.022343  0.143856 -0.057011  0.083098   \n",
       "1 -0.075250  0.195092  0.243829 -0.169353 -0.028060 -0.035800 -0.003912   \n",
       "2 -0.103785  0.226329  0.246068 -0.186726 -0.019194  0.057719  0.002511   \n",
       "3 -0.094587  0.260710  0.183705 -0.144703 -0.017368  0.005288  0.011188   \n",
       "4 -0.064907  0.225143  0.203515 -0.144751 -0.006472 -0.011114  0.036661   \n",
       "\n",
       "       4468      1012  \n",
       "0 -0.336973  0.044195  \n",
       "1 -0.216888  0.105810  \n",
       "2 -0.225501  0.102304  \n",
       "3 -0.212589  0.083279  \n",
       "4 -0.216034  0.083834  \n",
       "\n",
       "[5 rows x 115 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features3 = features[corr_columns_hemo]\n",
    "features3.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "9 peptides with predicted hemolysis% < 2%\n"
     ]
    }
   ],
   "source": [
    "# creates a df to save hemolysis score\n",
    "res_hemo = df.iloc[:,0:1].copy()\n",
    "\n",
    "# calculates the the hemolysis score\n",
    "pred= model_SVR_hemo.predict(features3)\n",
    "res_hemo['hemolysis_pred']= pred.round(2)\n",
    "\n",
    "#number of peptides with hemolysis score < 1\n",
    "print(f\"{res_hemo.query('hemolysis_pred <= 2').shape[0]} peptides with predicted hemolysis% < 2%\")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Results of single mutant"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>Sequence</th>\n",
       "      <th>MBC_8_prob</th>\n",
       "      <th>selectivity_pred</th>\n",
       "      <th>hemolysis_pred</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1mut_0</td>\n",
       "      <td>RGGRLCYCRRRFCVLVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.94</td>\n",
       "      <td>8.910000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1mut_1</td>\n",
       "      <td>RMGRLCYCRRRFCVCVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2.34</td>\n",
       "      <td>35.299999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1mut_2</td>\n",
       "      <td>RGGRLCYCRRRFCVCVFR</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.81</td>\n",
       "      <td>42.709999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1mut_3</td>\n",
       "      <td>EGGRLCYCRRRFCVCVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2.30</td>\n",
       "      <td>30.990000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1mut_4</td>\n",
       "      <td>RGGNLCYCRRRFCVCVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2.23</td>\n",
       "      <td>28.780001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>338</th>\n",
       "      <td>1mut_338</td>\n",
       "      <td>RGGRLCYSRRRFCVCVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2.72</td>\n",
       "      <td>15.270000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>339</th>\n",
       "      <td>1mut_339</td>\n",
       "      <td>RGGRLCYCRRKFCVCVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2.32</td>\n",
       "      <td>28.280001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>340</th>\n",
       "      <td>1mut_340</td>\n",
       "      <td>RGGRLCHCRRRFCVCVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.86</td>\n",
       "      <td>14.550000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>341</th>\n",
       "      <td>1mut_341</td>\n",
       "      <td>RGGRLCYCRRDFCVCVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2.17</td>\n",
       "      <td>17.010000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>342</th>\n",
       "      <td>1mut_342</td>\n",
       "      <td>RGGRLCYCRRRFPVCVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.37</td>\n",
       "      <td>4.620000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>343 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           ID            Sequence  MBC_8_prob  selectivity_pred  \\\n",
       "0      1mut_0  RGGRLCYCRRRFCVLVGR       100.0              1.94   \n",
       "1      1mut_1  RMGRLCYCRRRFCVCVGR       100.0              2.34   \n",
       "2      1mut_2  RGGRLCYCRRRFCVCVFR         0.0              2.81   \n",
       "3      1mut_3  EGGRLCYCRRRFCVCVGR       100.0              2.30   \n",
       "4      1mut_4  RGGNLCYCRRRFCVCVGR       100.0              2.23   \n",
       "..        ...                 ...         ...               ...   \n",
       "338  1mut_338  RGGRLCYSRRRFCVCVGR       100.0              2.72   \n",
       "339  1mut_339  RGGRLCYCRRKFCVCVGR       100.0              2.32   \n",
       "340  1mut_340  RGGRLCHCRRRFCVCVGR       100.0              1.86   \n",
       "341  1mut_341  RGGRLCYCRRDFCVCVGR       100.0              2.17   \n",
       "342  1mut_342  RGGRLCYCRRRFPVCVGR       100.0              1.37   \n",
       "\n",
       "     hemolysis_pred  \n",
       "0          8.910000  \n",
       "1         35.299999  \n",
       "2         42.709999  \n",
       "3         30.990000  \n",
       "4         28.780001  \n",
       "..              ...  \n",
       "338       15.270000  \n",
       "339       28.280001  \n",
       "340       14.550000  \n",
       "341       17.010000  \n",
       "342        4.620000  \n",
       "\n",
       "[343 rows x 5 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results = res_GNB.merge(res_sel[['ID', 'selectivity_pred']], how='inner', left_on='ID', right_on='ID')\n",
    "results = results.merge(res_hemo[['ID', 'hemolysis_pred']], how='inner', left_on='ID', right_on='ID')\n",
    "results = meta_data.merge(results, how='inner', left_on='ID', right_on='ID')\n",
    "results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Define the sets as lists of elements\n",
    "set1 = set(results.query('MBC_8_prob == 100')['ID'].values)\n",
    "set2 = set(results.query('selectivity_pred < 0.5')['ID'].values)\n",
    "set3 = set(results.query('hemolysis_pred <= 2')['ID'].values)\n",
    "\n",
    "\n",
    "# Create the Venn diagram\n",
    "venn3_unweighted([set1, set2, set3], set_labels=('Activity', 'Selectivity', 'Hemolysis'), set_colors=('b', 'g', 'r'))\n",
    "#plt.savefig('MBC_HEMO_SEL_2mut_venn3_predictions_v03.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# saving results\n",
    "results.to_csv('results/predictions_MIC_SEL_HEMO_models_PG1_1mut_v01.2.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Predictions on double mutation Protegrin-1 Variants"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>Sequence</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2mut_0</td>\n",
       "      <td>RGPRLCYCRRRQCVCVGR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2mut_1</td>\n",
       "      <td>YGGPLCYCRRRFCVCVGR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2mut_2</td>\n",
       "      <td>RKGRLCGCRRRFCVCVGR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2mut_3</td>\n",
       "      <td>RGGRLCYCRRWFCVKVGR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2mut_4</td>\n",
       "      <td>RGGRLCYCMRRFCVCDGR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55571</th>\n",
       "      <td>2mut_55571</td>\n",
       "      <td>CGGRLCYCRRRWCVCVGR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55572</th>\n",
       "      <td>2mut_55572</td>\n",
       "      <td>RGGRLCYCRRRHDVCVGR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55573</th>\n",
       "      <td>2mut_55573</td>\n",
       "      <td>RGGRMCYCRRRFCVCVAR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55574</th>\n",
       "      <td>2mut_55574</td>\n",
       "      <td>RGGRLCYCRIRFIVCVGR</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55575</th>\n",
       "      <td>2mut_55575</td>\n",
       "      <td>RGGRLCYWRRRFCVCVGP</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>55576 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               ID            Sequence\n",
       "0          2mut_0  RGPRLCYCRRRQCVCVGR\n",
       "1          2mut_1  YGGPLCYCRRRFCVCVGR\n",
       "2          2mut_2  RKGRLCGCRRRFCVCVGR\n",
       "3          2mut_3  RGGRLCYCRRWFCVKVGR\n",
       "4          2mut_4  RGGRLCYCMRRFCVCDGR\n",
       "...           ...                 ...\n",
       "55571  2mut_55571  CGGRLCYCRRRWCVCVGR\n",
       "55572  2mut_55572  RGGRLCYCRRRHDVCVGR\n",
       "55573  2mut_55573  RGGRMCYCRRRFCVCVAR\n",
       "55574  2mut_55574  RGGRLCYCRIRFIVCVGR\n",
       "55575  2mut_55575  RGGRLCYWRRRFCVCVGP\n",
       "\n",
       "[55576 rows x 2 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load meta_data with mutations on PG1\n",
    "meta_data = pd.read_csv('data/protegrin_variants_2mut.csv', index_col=0)\n",
    "meta_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "######################## Load the embedding ######################\n",
    "# since those embeddings are too large, I saved them in a separate file\n",
    "\n",
    "# dir_path = '/stor/work/Wilke/luiz/ML_PG1/embeddings/esm2_15B_pg1_2mut/'\n",
    "# embeddings = {}\n",
    "# for file in os.listdir(dir_path):\n",
    "#         file_path = str(dir_path) + file \n",
    "#         label = file.split('.pt')[0]\n",
    "#         embeddings[label] = np.array(torch.load(file_path)['mean_representations'][48])\n",
    "\n",
    "\n",
    "# # merge embeddings with metadata\n",
    "# df = meta_data.merge(pd.DataFrame.from_dict(embeddings, orient='index'), how='inner', left_on='ID', right_index=True)\n",
    "\n",
    "# df.to_pickle('data/embedding/esm2_15B_pg1_2mut.pkl')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>Sequence</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "      <th>17</th>\n",
       "      <th>18</th>\n",
       "      <th>19</th>\n",
       "      <th>20</th>\n",
       "      <th>21</th>\n",
       "      <th>22</th>\n",
       "      <th>23</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "      <th>26</th>\n",
       "      <th>27</th>\n",
       "      <th>28</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "      <th>33</th>\n",
       "      <th>34</th>\n",
       "      <th>35</th>\n",
       "      <th>36</th>\n",
       "      <th>37</th>\n",
       "      <th>38</th>\n",
       "      <th>39</th>\n",
       "      <th>40</th>\n",
       "      <th>41</th>\n",
       "      <th>42</th>\n",
       "      <th>43</th>\n",
       "      <th>44</th>\n",
       "      <th>45</th>\n",
       "      <th>46</th>\n",
       "      <th>47</th>\n",
       "      <th>...</th>\n",
       "      <th>5070</th>\n",
       "      <th>5071</th>\n",
       "      <th>5072</th>\n",
       "      <th>5073</th>\n",
       "      <th>5074</th>\n",
       "      <th>5075</th>\n",
       "      <th>5076</th>\n",
       "      <th>5077</th>\n",
       "      <th>5078</th>\n",
       "      <th>5079</th>\n",
       "      <th>5080</th>\n",
       "      <th>5081</th>\n",
       "      <th>5082</th>\n",
       "      <th>5083</th>\n",
       "      <th>5084</th>\n",
       "      <th>5085</th>\n",
       "      <th>5086</th>\n",
       "      <th>5087</th>\n",
       "      <th>5088</th>\n",
       "      <th>5089</th>\n",
       "      <th>5090</th>\n",
       "      <th>5091</th>\n",
       "      <th>5092</th>\n",
       "      <th>5093</th>\n",
       "      <th>5094</th>\n",
       "      <th>5095</th>\n",
       "      <th>5096</th>\n",
       "      <th>5097</th>\n",
       "      <th>5098</th>\n",
       "      <th>5099</th>\n",
       "      <th>5100</th>\n",
       "      <th>5101</th>\n",
       "      <th>5102</th>\n",
       "      <th>5103</th>\n",
       "      <th>5104</th>\n",
       "      <th>5105</th>\n",
       "      <th>5106</th>\n",
       "      <th>5107</th>\n",
       "      <th>5108</th>\n",
       "      <th>5109</th>\n",
       "      <th>5110</th>\n",
       "      <th>5111</th>\n",
       "      <th>5112</th>\n",
       "      <th>5113</th>\n",
       "      <th>5114</th>\n",
       "      <th>5115</th>\n",
       "      <th>5116</th>\n",
       "      <th>5117</th>\n",
       "      <th>5118</th>\n",
       "      <th>5119</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2mut_0</td>\n",
       "      <td>RGPRLCYCRRRQCVCVGR</td>\n",
       "      <td>0.030471</td>\n",
       "      <td>-0.036276</td>\n",
       "      <td>-0.111070</td>\n",
       "      <td>0.150194</td>\n",
       "      <td>-0.230985</td>\n",
       "      <td>0.181216</td>\n",
       "      <td>-0.178246</td>\n",
       "      <td>0.158280</td>\n",
       "      <td>-0.044290</td>\n",
       "      <td>0.140545</td>\n",
       "      <td>-0.173849</td>\n",
       "      <td>-0.074231</td>\n",
       "      <td>0.127456</td>\n",
       "      <td>0.139890</td>\n",
       "      <td>0.141800</td>\n",
       "      <td>-0.161217</td>\n",
       "      <td>0.052185</td>\n",
       "      <td>-0.073374</td>\n",
       "      <td>0.231049</td>\n",
       "      <td>0.004841</td>\n",
       "      <td>0.076786</td>\n",
       "      <td>-0.000564</td>\n",
       "      <td>-0.235507</td>\n",
       "      <td>0.093585</td>\n",
       "      <td>-0.094216</td>\n",
       "      <td>-0.030184</td>\n",
       "      <td>0.090808</td>\n",
       "      <td>-0.017730</td>\n",
       "      <td>0.056953</td>\n",
       "      <td>0.168276</td>\n",
       "      <td>-0.138922</td>\n",
       "      <td>-0.055812</td>\n",
       "      <td>-0.238720</td>\n",
       "      <td>-0.121912</td>\n",
       "      <td>-0.131041</td>\n",
       "      <td>0.009613</td>\n",
       "      <td>0.166661</td>\n",
       "      <td>-0.117495</td>\n",
       "      <td>0.190636</td>\n",
       "      <td>-0.058775</td>\n",
       "      <td>0.083950</td>\n",
       "      <td>-0.137385</td>\n",
       "      <td>0.066807</td>\n",
       "      <td>0.088748</td>\n",
       "      <td>-0.054584</td>\n",
       "      <td>0.269736</td>\n",
       "      <td>0.049476</td>\n",
       "      <td>0.108028</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.332250</td>\n",
       "      <td>-0.182212</td>\n",
       "      <td>0.188159</td>\n",
       "      <td>-0.089034</td>\n",
       "      <td>-0.064079</td>\n",
       "      <td>0.156366</td>\n",
       "      <td>-0.090362</td>\n",
       "      <td>-0.010652</td>\n",
       "      <td>0.187834</td>\n",
       "      <td>0.213740</td>\n",
       "      <td>-0.055363</td>\n",
       "      <td>-0.009788</td>\n",
       "      <td>0.014521</td>\n",
       "      <td>0.185780</td>\n",
       "      <td>0.124452</td>\n",
       "      <td>0.013869</td>\n",
       "      <td>-0.185609</td>\n",
       "      <td>0.156731</td>\n",
       "      <td>-0.108033</td>\n",
       "      <td>-0.137903</td>\n",
       "      <td>-0.097508</td>\n",
       "      <td>-0.199132</td>\n",
       "      <td>-0.055758</td>\n",
       "      <td>-0.013001</td>\n",
       "      <td>0.282426</td>\n",
       "      <td>-0.031745</td>\n",
       "      <td>0.175234</td>\n",
       "      <td>-0.038487</td>\n",
       "      <td>0.071631</td>\n",
       "      <td>0.056087</td>\n",
       "      <td>0.094258</td>\n",
       "      <td>0.134074</td>\n",
       "      <td>-0.014872</td>\n",
       "      <td>-0.083226</td>\n",
       "      <td>-0.015783</td>\n",
       "      <td>0.033664</td>\n",
       "      <td>-0.126464</td>\n",
       "      <td>-0.285006</td>\n",
       "      <td>-0.048229</td>\n",
       "      <td>0.061758</td>\n",
       "      <td>-0.191082</td>\n",
       "      <td>-0.151428</td>\n",
       "      <td>0.059017</td>\n",
       "      <td>-0.244604</td>\n",
       "      <td>0.117419</td>\n",
       "      <td>-0.155859</td>\n",
       "      <td>-0.045599</td>\n",
       "      <td>-0.075758</td>\n",
       "      <td>-0.259623</td>\n",
       "      <td>0.045590</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2mut_1</td>\n",
       "      <td>YGGPLCYCRRRFCVCVGR</td>\n",
       "      <td>0.078283</td>\n",
       "      <td>-0.004239</td>\n",
       "      <td>-0.083609</td>\n",
       "      <td>0.105327</td>\n",
       "      <td>-0.233435</td>\n",
       "      <td>0.161460</td>\n",
       "      <td>-0.106038</td>\n",
       "      <td>0.203431</td>\n",
       "      <td>-0.081741</td>\n",
       "      <td>0.155934</td>\n",
       "      <td>-0.148036</td>\n",
       "      <td>-0.138177</td>\n",
       "      <td>0.127338</td>\n",
       "      <td>0.172699</td>\n",
       "      <td>0.143295</td>\n",
       "      <td>-0.190010</td>\n",
       "      <td>0.101302</td>\n",
       "      <td>-0.077433</td>\n",
       "      <td>0.149300</td>\n",
       "      <td>0.018422</td>\n",
       "      <td>0.155382</td>\n",
       "      <td>0.059158</td>\n",
       "      <td>-0.202118</td>\n",
       "      <td>0.056027</td>\n",
       "      <td>-0.144371</td>\n",
       "      <td>-0.061357</td>\n",
       "      <td>0.085607</td>\n",
       "      <td>-0.040719</td>\n",
       "      <td>0.005678</td>\n",
       "      <td>0.153111</td>\n",
       "      <td>-0.134872</td>\n",
       "      <td>-0.038001</td>\n",
       "      <td>-0.213945</td>\n",
       "      <td>-0.130888</td>\n",
       "      <td>-0.058073</td>\n",
       "      <td>0.013071</td>\n",
       "      <td>0.204674</td>\n",
       "      <td>-0.049626</td>\n",
       "      <td>0.176748</td>\n",
       "      <td>-0.075964</td>\n",
       "      <td>0.177314</td>\n",
       "      <td>-0.164634</td>\n",
       "      <td>0.122157</td>\n",
       "      <td>0.013722</td>\n",
       "      <td>-0.078917</td>\n",
       "      <td>0.326256</td>\n",
       "      <td>0.102039</td>\n",
       "      <td>0.064966</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.208491</td>\n",
       "      <td>-0.181908</td>\n",
       "      <td>0.204974</td>\n",
       "      <td>0.013904</td>\n",
       "      <td>-0.081345</td>\n",
       "      <td>0.084125</td>\n",
       "      <td>-0.008529</td>\n",
       "      <td>-0.013594</td>\n",
       "      <td>0.245070</td>\n",
       "      <td>0.272875</td>\n",
       "      <td>-0.054534</td>\n",
       "      <td>0.077179</td>\n",
       "      <td>0.028892</td>\n",
       "      <td>0.122911</td>\n",
       "      <td>0.150612</td>\n",
       "      <td>-0.056876</td>\n",
       "      <td>-0.118520</td>\n",
       "      <td>0.162811</td>\n",
       "      <td>-0.106010</td>\n",
       "      <td>-0.062028</td>\n",
       "      <td>-0.084144</td>\n",
       "      <td>-0.140908</td>\n",
       "      <td>-0.094160</td>\n",
       "      <td>-0.074398</td>\n",
       "      <td>0.280061</td>\n",
       "      <td>-0.020089</td>\n",
       "      <td>0.090262</td>\n",
       "      <td>0.008972</td>\n",
       "      <td>-0.009543</td>\n",
       "      <td>0.066797</td>\n",
       "      <td>0.110991</td>\n",
       "      <td>0.168301</td>\n",
       "      <td>-0.014092</td>\n",
       "      <td>-0.092231</td>\n",
       "      <td>0.006551</td>\n",
       "      <td>0.018966</td>\n",
       "      <td>-0.110038</td>\n",
       "      <td>-0.285004</td>\n",
       "      <td>-0.044679</td>\n",
       "      <td>0.036731</td>\n",
       "      <td>-0.182719</td>\n",
       "      <td>-0.162484</td>\n",
       "      <td>0.037435</td>\n",
       "      <td>-0.229143</td>\n",
       "      <td>0.173071</td>\n",
       "      <td>-0.190800</td>\n",
       "      <td>-0.026766</td>\n",
       "      <td>-0.103359</td>\n",
       "      <td>-0.266636</td>\n",
       "      <td>0.025352</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2mut_2</td>\n",
       "      <td>RKGRLCGCRRRFCVCVGR</td>\n",
       "      <td>0.085428</td>\n",
       "      <td>-0.057224</td>\n",
       "      <td>-0.048115</td>\n",
       "      <td>0.131888</td>\n",
       "      <td>-0.233556</td>\n",
       "      <td>0.146629</td>\n",
       "      <td>-0.058954</td>\n",
       "      <td>0.145036</td>\n",
       "      <td>-0.012269</td>\n",
       "      <td>0.208369</td>\n",
       "      <td>-0.116377</td>\n",
       "      <td>-0.063401</td>\n",
       "      <td>0.179298</td>\n",
       "      <td>0.176179</td>\n",
       "      <td>0.089100</td>\n",
       "      <td>-0.128998</td>\n",
       "      <td>0.192588</td>\n",
       "      <td>-0.062569</td>\n",
       "      <td>0.056391</td>\n",
       "      <td>-0.061444</td>\n",
       "      <td>0.136877</td>\n",
       "      <td>0.051762</td>\n",
       "      <td>-0.240653</td>\n",
       "      <td>0.131735</td>\n",
       "      <td>-0.155009</td>\n",
       "      <td>-0.012141</td>\n",
       "      <td>0.110780</td>\n",
       "      <td>-0.081294</td>\n",
       "      <td>0.068365</td>\n",
       "      <td>0.167697</td>\n",
       "      <td>-0.143250</td>\n",
       "      <td>0.005252</td>\n",
       "      <td>-0.351505</td>\n",
       "      <td>-0.103353</td>\n",
       "      <td>-0.154235</td>\n",
       "      <td>0.003228</td>\n",
       "      <td>0.185293</td>\n",
       "      <td>-0.056369</td>\n",
       "      <td>0.242288</td>\n",
       "      <td>-0.047226</td>\n",
       "      <td>0.159304</td>\n",
       "      <td>-0.101366</td>\n",
       "      <td>0.098639</td>\n",
       "      <td>0.058434</td>\n",
       "      <td>-0.045043</td>\n",
       "      <td>0.278100</td>\n",
       "      <td>0.043613</td>\n",
       "      <td>0.063437</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.328105</td>\n",
       "      <td>-0.162223</td>\n",
       "      <td>0.256969</td>\n",
       "      <td>-0.061340</td>\n",
       "      <td>-0.029240</td>\n",
       "      <td>0.072979</td>\n",
       "      <td>-0.043325</td>\n",
       "      <td>-0.013580</td>\n",
       "      <td>0.229346</td>\n",
       "      <td>0.264277</td>\n",
       "      <td>-0.033285</td>\n",
       "      <td>0.013914</td>\n",
       "      <td>-0.017617</td>\n",
       "      <td>0.160241</td>\n",
       "      <td>0.176237</td>\n",
       "      <td>-0.087623</td>\n",
       "      <td>-0.177227</td>\n",
       "      <td>0.208036</td>\n",
       "      <td>-0.145973</td>\n",
       "      <td>0.000206</td>\n",
       "      <td>-0.188131</td>\n",
       "      <td>-0.057958</td>\n",
       "      <td>-0.108080</td>\n",
       "      <td>-0.064826</td>\n",
       "      <td>0.241150</td>\n",
       "      <td>0.070687</td>\n",
       "      <td>0.167148</td>\n",
       "      <td>-0.012571</td>\n",
       "      <td>-0.008496</td>\n",
       "      <td>0.036840</td>\n",
       "      <td>0.092076</td>\n",
       "      <td>0.183831</td>\n",
       "      <td>-0.038787</td>\n",
       "      <td>-0.071339</td>\n",
       "      <td>-0.011860</td>\n",
       "      <td>-0.058060</td>\n",
       "      <td>-0.045512</td>\n",
       "      <td>-0.228114</td>\n",
       "      <td>-0.049612</td>\n",
       "      <td>-0.007920</td>\n",
       "      <td>-0.166855</td>\n",
       "      <td>-0.198541</td>\n",
       "      <td>0.027955</td>\n",
       "      <td>-0.190623</td>\n",
       "      <td>0.077585</td>\n",
       "      <td>-0.172339</td>\n",
       "      <td>0.024809</td>\n",
       "      <td>-0.092293</td>\n",
       "      <td>-0.187360</td>\n",
       "      <td>0.020935</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2mut_3</td>\n",
       "      <td>RGGRLCYCRRWFCVKVGR</td>\n",
       "      <td>0.057880</td>\n",
       "      <td>-0.050223</td>\n",
       "      <td>-0.124206</td>\n",
       "      <td>0.076189</td>\n",
       "      <td>-0.272554</td>\n",
       "      <td>0.140542</td>\n",
       "      <td>-0.079627</td>\n",
       "      <td>0.214544</td>\n",
       "      <td>-0.021029</td>\n",
       "      <td>0.190860</td>\n",
       "      <td>-0.159807</td>\n",
       "      <td>-0.098083</td>\n",
       "      <td>0.199157</td>\n",
       "      <td>0.179230</td>\n",
       "      <td>0.109095</td>\n",
       "      <td>-0.135120</td>\n",
       "      <td>0.113276</td>\n",
       "      <td>-0.068056</td>\n",
       "      <td>0.066501</td>\n",
       "      <td>0.022015</td>\n",
       "      <td>0.165217</td>\n",
       "      <td>0.009979</td>\n",
       "      <td>-0.291171</td>\n",
       "      <td>0.147817</td>\n",
       "      <td>-0.147829</td>\n",
       "      <td>-0.041164</td>\n",
       "      <td>0.068331</td>\n",
       "      <td>-0.102698</td>\n",
       "      <td>-0.041732</td>\n",
       "      <td>0.109131</td>\n",
       "      <td>-0.132808</td>\n",
       "      <td>-0.042759</td>\n",
       "      <td>-0.286052</td>\n",
       "      <td>-0.113242</td>\n",
       "      <td>-0.090135</td>\n",
       "      <td>-0.022518</td>\n",
       "      <td>0.168574</td>\n",
       "      <td>-0.058750</td>\n",
       "      <td>0.236315</td>\n",
       "      <td>-0.028953</td>\n",
       "      <td>0.159713</td>\n",
       "      <td>-0.287824</td>\n",
       "      <td>0.157747</td>\n",
       "      <td>0.060543</td>\n",
       "      <td>-0.057961</td>\n",
       "      <td>0.306598</td>\n",
       "      <td>0.056881</td>\n",
       "      <td>0.039657</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.307967</td>\n",
       "      <td>-0.126027</td>\n",
       "      <td>0.237130</td>\n",
       "      <td>-0.060796</td>\n",
       "      <td>-0.106810</td>\n",
       "      <td>0.076391</td>\n",
       "      <td>-0.032235</td>\n",
       "      <td>0.007105</td>\n",
       "      <td>0.312369</td>\n",
       "      <td>0.172563</td>\n",
       "      <td>-0.045592</td>\n",
       "      <td>0.099534</td>\n",
       "      <td>0.065470</td>\n",
       "      <td>0.134455</td>\n",
       "      <td>0.143874</td>\n",
       "      <td>0.002906</td>\n",
       "      <td>-0.142259</td>\n",
       "      <td>0.190600</td>\n",
       "      <td>-0.021462</td>\n",
       "      <td>-0.018197</td>\n",
       "      <td>-0.208055</td>\n",
       "      <td>-0.071023</td>\n",
       "      <td>-0.081870</td>\n",
       "      <td>-0.171144</td>\n",
       "      <td>0.213066</td>\n",
       "      <td>-0.009455</td>\n",
       "      <td>0.160588</td>\n",
       "      <td>-0.060975</td>\n",
       "      <td>-0.063147</td>\n",
       "      <td>0.025246</td>\n",
       "      <td>0.081391</td>\n",
       "      <td>0.122101</td>\n",
       "      <td>-0.058024</td>\n",
       "      <td>-0.072349</td>\n",
       "      <td>-0.008517</td>\n",
       "      <td>-0.108182</td>\n",
       "      <td>-0.060309</td>\n",
       "      <td>-0.290993</td>\n",
       "      <td>-0.001576</td>\n",
       "      <td>0.017886</td>\n",
       "      <td>-0.120012</td>\n",
       "      <td>-0.084091</td>\n",
       "      <td>0.096398</td>\n",
       "      <td>-0.096984</td>\n",
       "      <td>0.099995</td>\n",
       "      <td>-0.157944</td>\n",
       "      <td>-0.051621</td>\n",
       "      <td>-0.029807</td>\n",
       "      <td>-0.291391</td>\n",
       "      <td>0.005705</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2mut_4</td>\n",
       "      <td>RGGRLCYCMRRFCVCDGR</td>\n",
       "      <td>0.212485</td>\n",
       "      <td>0.024579</td>\n",
       "      <td>-0.084416</td>\n",
       "      <td>0.144025</td>\n",
       "      <td>-0.232260</td>\n",
       "      <td>0.099251</td>\n",
       "      <td>-0.095035</td>\n",
       "      <td>0.196182</td>\n",
       "      <td>-0.108374</td>\n",
       "      <td>0.186943</td>\n",
       "      <td>-0.141637</td>\n",
       "      <td>-0.176585</td>\n",
       "      <td>0.178890</td>\n",
       "      <td>0.219366</td>\n",
       "      <td>0.157616</td>\n",
       "      <td>-0.134976</td>\n",
       "      <td>0.083950</td>\n",
       "      <td>-0.133882</td>\n",
       "      <td>0.113217</td>\n",
       "      <td>0.087540</td>\n",
       "      <td>0.160665</td>\n",
       "      <td>0.063379</td>\n",
       "      <td>-0.183397</td>\n",
       "      <td>0.080644</td>\n",
       "      <td>-0.179947</td>\n",
       "      <td>-0.104130</td>\n",
       "      <td>0.103943</td>\n",
       "      <td>-0.012196</td>\n",
       "      <td>0.027808</td>\n",
       "      <td>0.159193</td>\n",
       "      <td>-0.135980</td>\n",
       "      <td>-0.071504</td>\n",
       "      <td>-0.267629</td>\n",
       "      <td>-0.091737</td>\n",
       "      <td>-0.073513</td>\n",
       "      <td>0.039947</td>\n",
       "      <td>0.202282</td>\n",
       "      <td>-0.122841</td>\n",
       "      <td>0.191903</td>\n",
       "      <td>-0.085785</td>\n",
       "      <td>0.163927</td>\n",
       "      <td>-0.266495</td>\n",
       "      <td>0.073727</td>\n",
       "      <td>0.059343</td>\n",
       "      <td>-0.140880</td>\n",
       "      <td>0.346234</td>\n",
       "      <td>0.095565</td>\n",
       "      <td>0.073925</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.249055</td>\n",
       "      <td>-0.163885</td>\n",
       "      <td>0.184048</td>\n",
       "      <td>-0.029923</td>\n",
       "      <td>0.018876</td>\n",
       "      <td>0.082031</td>\n",
       "      <td>-0.042319</td>\n",
       "      <td>0.006107</td>\n",
       "      <td>0.219458</td>\n",
       "      <td>0.257496</td>\n",
       "      <td>-0.044507</td>\n",
       "      <td>0.121740</td>\n",
       "      <td>0.041201</td>\n",
       "      <td>0.099270</td>\n",
       "      <td>0.136533</td>\n",
       "      <td>-0.048466</td>\n",
       "      <td>-0.100322</td>\n",
       "      <td>0.195729</td>\n",
       "      <td>-0.145436</td>\n",
       "      <td>-0.033395</td>\n",
       "      <td>-0.071495</td>\n",
       "      <td>-0.146533</td>\n",
       "      <td>-0.127662</td>\n",
       "      <td>-0.067469</td>\n",
       "      <td>0.307449</td>\n",
       "      <td>-0.044970</td>\n",
       "      <td>0.040246</td>\n",
       "      <td>-0.006643</td>\n",
       "      <td>-0.103936</td>\n",
       "      <td>0.064969</td>\n",
       "      <td>0.220303</td>\n",
       "      <td>0.141593</td>\n",
       "      <td>-0.029952</td>\n",
       "      <td>-0.114162</td>\n",
       "      <td>-0.001476</td>\n",
       "      <td>-0.008912</td>\n",
       "      <td>-0.056513</td>\n",
       "      <td>-0.308030</td>\n",
       "      <td>0.002654</td>\n",
       "      <td>-0.009495</td>\n",
       "      <td>-0.175722</td>\n",
       "      <td>-0.094956</td>\n",
       "      <td>0.136943</td>\n",
       "      <td>-0.242696</td>\n",
       "      <td>0.129612</td>\n",
       "      <td>-0.228823</td>\n",
       "      <td>-0.060089</td>\n",
       "      <td>-0.118763</td>\n",
       "      <td>-0.227903</td>\n",
       "      <td>0.101617</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55571</th>\n",
       "      <td>2mut_55571</td>\n",
       "      <td>CGGRLCYCRRRWCVCVGR</td>\n",
       "      <td>0.082458</td>\n",
       "      <td>-0.054124</td>\n",
       "      <td>-0.153812</td>\n",
       "      <td>0.028242</td>\n",
       "      <td>-0.250708</td>\n",
       "      <td>0.075041</td>\n",
       "      <td>-0.087562</td>\n",
       "      <td>0.206055</td>\n",
       "      <td>0.005263</td>\n",
       "      <td>0.066554</td>\n",
       "      <td>-0.187047</td>\n",
       "      <td>-0.055146</td>\n",
       "      <td>0.113620</td>\n",
       "      <td>0.230673</td>\n",
       "      <td>0.062871</td>\n",
       "      <td>-0.091331</td>\n",
       "      <td>0.158153</td>\n",
       "      <td>-0.072747</td>\n",
       "      <td>0.209477</td>\n",
       "      <td>0.075546</td>\n",
       "      <td>0.100096</td>\n",
       "      <td>-0.018672</td>\n",
       "      <td>-0.173173</td>\n",
       "      <td>0.077043</td>\n",
       "      <td>-0.020988</td>\n",
       "      <td>-0.079208</td>\n",
       "      <td>0.131811</td>\n",
       "      <td>0.014909</td>\n",
       "      <td>0.012486</td>\n",
       "      <td>0.175663</td>\n",
       "      <td>-0.191909</td>\n",
       "      <td>0.008557</td>\n",
       "      <td>-0.159104</td>\n",
       "      <td>-0.107365</td>\n",
       "      <td>-0.112420</td>\n",
       "      <td>0.019998</td>\n",
       "      <td>0.171493</td>\n",
       "      <td>-0.049025</td>\n",
       "      <td>0.061186</td>\n",
       "      <td>-0.171707</td>\n",
       "      <td>0.127845</td>\n",
       "      <td>-0.080232</td>\n",
       "      <td>0.065033</td>\n",
       "      <td>0.062051</td>\n",
       "      <td>-0.077829</td>\n",
       "      <td>0.249108</td>\n",
       "      <td>0.134323</td>\n",
       "      <td>0.107528</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.235606</td>\n",
       "      <td>-0.205449</td>\n",
       "      <td>0.146447</td>\n",
       "      <td>-0.050081</td>\n",
       "      <td>-0.064856</td>\n",
       "      <td>0.077309</td>\n",
       "      <td>-0.010093</td>\n",
       "      <td>0.033059</td>\n",
       "      <td>0.178510</td>\n",
       "      <td>0.267095</td>\n",
       "      <td>-0.008456</td>\n",
       "      <td>0.099262</td>\n",
       "      <td>0.064182</td>\n",
       "      <td>0.086947</td>\n",
       "      <td>0.210259</td>\n",
       "      <td>-0.014526</td>\n",
       "      <td>-0.158287</td>\n",
       "      <td>0.158621</td>\n",
       "      <td>-0.075987</td>\n",
       "      <td>-0.103462</td>\n",
       "      <td>-0.080306</td>\n",
       "      <td>-0.153473</td>\n",
       "      <td>-0.079218</td>\n",
       "      <td>0.015159</td>\n",
       "      <td>0.188657</td>\n",
       "      <td>-0.095693</td>\n",
       "      <td>0.113370</td>\n",
       "      <td>-0.031929</td>\n",
       "      <td>0.017222</td>\n",
       "      <td>0.075270</td>\n",
       "      <td>0.031181</td>\n",
       "      <td>0.160164</td>\n",
       "      <td>0.013887</td>\n",
       "      <td>-0.065848</td>\n",
       "      <td>0.062114</td>\n",
       "      <td>0.035505</td>\n",
       "      <td>-0.174763</td>\n",
       "      <td>-0.289324</td>\n",
       "      <td>-0.014531</td>\n",
       "      <td>-0.020021</td>\n",
       "      <td>-0.158223</td>\n",
       "      <td>-0.175883</td>\n",
       "      <td>0.036970</td>\n",
       "      <td>-0.238702</td>\n",
       "      <td>0.154410</td>\n",
       "      <td>-0.193272</td>\n",
       "      <td>-0.014171</td>\n",
       "      <td>-0.038470</td>\n",
       "      <td>-0.281103</td>\n",
       "      <td>0.036803</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55572</th>\n",
       "      <td>2mut_55572</td>\n",
       "      <td>RGGRLCYCRRRHDVCVGR</td>\n",
       "      <td>0.005155</td>\n",
       "      <td>-0.164447</td>\n",
       "      <td>-0.016448</td>\n",
       "      <td>0.263290</td>\n",
       "      <td>-0.190351</td>\n",
       "      <td>0.191838</td>\n",
       "      <td>-0.132956</td>\n",
       "      <td>0.206665</td>\n",
       "      <td>-0.053722</td>\n",
       "      <td>0.319446</td>\n",
       "      <td>-0.217972</td>\n",
       "      <td>-0.144265</td>\n",
       "      <td>0.222120</td>\n",
       "      <td>0.061943</td>\n",
       "      <td>0.056246</td>\n",
       "      <td>-0.118007</td>\n",
       "      <td>-0.077944</td>\n",
       "      <td>-0.173787</td>\n",
       "      <td>0.159124</td>\n",
       "      <td>0.045858</td>\n",
       "      <td>0.058331</td>\n",
       "      <td>0.020416</td>\n",
       "      <td>-0.504820</td>\n",
       "      <td>0.120175</td>\n",
       "      <td>-0.255182</td>\n",
       "      <td>-0.000031</td>\n",
       "      <td>0.096224</td>\n",
       "      <td>-0.195531</td>\n",
       "      <td>-0.043021</td>\n",
       "      <td>0.209337</td>\n",
       "      <td>-0.070500</td>\n",
       "      <td>0.015667</td>\n",
       "      <td>-0.349645</td>\n",
       "      <td>-0.097594</td>\n",
       "      <td>-0.057167</td>\n",
       "      <td>0.002473</td>\n",
       "      <td>0.098192</td>\n",
       "      <td>-0.053422</td>\n",
       "      <td>0.365413</td>\n",
       "      <td>0.010262</td>\n",
       "      <td>0.145131</td>\n",
       "      <td>-0.265709</td>\n",
       "      <td>0.127403</td>\n",
       "      <td>0.157150</td>\n",
       "      <td>-0.059583</td>\n",
       "      <td>0.351660</td>\n",
       "      <td>-0.118701</td>\n",
       "      <td>0.047485</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.184939</td>\n",
       "      <td>-0.108122</td>\n",
       "      <td>0.187322</td>\n",
       "      <td>-0.225827</td>\n",
       "      <td>0.000120</td>\n",
       "      <td>0.062208</td>\n",
       "      <td>-0.069744</td>\n",
       "      <td>0.117765</td>\n",
       "      <td>0.194843</td>\n",
       "      <td>0.217680</td>\n",
       "      <td>-0.024724</td>\n",
       "      <td>-0.040571</td>\n",
       "      <td>0.022824</td>\n",
       "      <td>0.105905</td>\n",
       "      <td>-0.018251</td>\n",
       "      <td>-0.013676</td>\n",
       "      <td>-0.171608</td>\n",
       "      <td>0.225520</td>\n",
       "      <td>-0.070250</td>\n",
       "      <td>0.049515</td>\n",
       "      <td>-0.161149</td>\n",
       "      <td>0.060027</td>\n",
       "      <td>-0.003002</td>\n",
       "      <td>-0.206381</td>\n",
       "      <td>0.318415</td>\n",
       "      <td>0.002611</td>\n",
       "      <td>0.147094</td>\n",
       "      <td>-0.064504</td>\n",
       "      <td>0.063217</td>\n",
       "      <td>-0.062706</td>\n",
       "      <td>0.111827</td>\n",
       "      <td>0.105259</td>\n",
       "      <td>-0.136614</td>\n",
       "      <td>-0.100203</td>\n",
       "      <td>-0.238890</td>\n",
       "      <td>-0.075930</td>\n",
       "      <td>0.019882</td>\n",
       "      <td>-0.311911</td>\n",
       "      <td>-0.020147</td>\n",
       "      <td>0.041474</td>\n",
       "      <td>-0.139836</td>\n",
       "      <td>-0.138856</td>\n",
       "      <td>0.037153</td>\n",
       "      <td>-0.283946</td>\n",
       "      <td>0.093709</td>\n",
       "      <td>-0.093828</td>\n",
       "      <td>0.197921</td>\n",
       "      <td>0.061686</td>\n",
       "      <td>-0.203528</td>\n",
       "      <td>0.005697</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55573</th>\n",
       "      <td>2mut_55573</td>\n",
       "      <td>RGGRMCYCRRRFCVCVAR</td>\n",
       "      <td>0.152141</td>\n",
       "      <td>0.008183</td>\n",
       "      <td>-0.151269</td>\n",
       "      <td>0.136687</td>\n",
       "      <td>-0.251317</td>\n",
       "      <td>0.134183</td>\n",
       "      <td>-0.108524</td>\n",
       "      <td>0.154696</td>\n",
       "      <td>-0.015930</td>\n",
       "      <td>0.115763</td>\n",
       "      <td>-0.151316</td>\n",
       "      <td>-0.068709</td>\n",
       "      <td>0.151451</td>\n",
       "      <td>0.204528</td>\n",
       "      <td>0.095688</td>\n",
       "      <td>-0.143107</td>\n",
       "      <td>0.124312</td>\n",
       "      <td>-0.089717</td>\n",
       "      <td>0.142609</td>\n",
       "      <td>0.001460</td>\n",
       "      <td>0.095826</td>\n",
       "      <td>-0.040184</td>\n",
       "      <td>-0.149959</td>\n",
       "      <td>-0.016682</td>\n",
       "      <td>-0.128044</td>\n",
       "      <td>0.004092</td>\n",
       "      <td>0.141307</td>\n",
       "      <td>-0.062854</td>\n",
       "      <td>0.089832</td>\n",
       "      <td>0.129261</td>\n",
       "      <td>-0.129841</td>\n",
       "      <td>-0.010646</td>\n",
       "      <td>-0.234781</td>\n",
       "      <td>-0.093643</td>\n",
       "      <td>-0.102319</td>\n",
       "      <td>-0.011205</td>\n",
       "      <td>0.151069</td>\n",
       "      <td>-0.078129</td>\n",
       "      <td>0.122194</td>\n",
       "      <td>-0.115878</td>\n",
       "      <td>0.146669</td>\n",
       "      <td>-0.119630</td>\n",
       "      <td>0.065052</td>\n",
       "      <td>0.037127</td>\n",
       "      <td>-0.081287</td>\n",
       "      <td>0.276126</td>\n",
       "      <td>0.145157</td>\n",
       "      <td>0.107750</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.313060</td>\n",
       "      <td>-0.161855</td>\n",
       "      <td>0.131321</td>\n",
       "      <td>-0.040697</td>\n",
       "      <td>-0.016497</td>\n",
       "      <td>0.123411</td>\n",
       "      <td>0.000906</td>\n",
       "      <td>0.016701</td>\n",
       "      <td>0.210516</td>\n",
       "      <td>0.298393</td>\n",
       "      <td>-0.057488</td>\n",
       "      <td>0.105326</td>\n",
       "      <td>0.011940</td>\n",
       "      <td>0.097368</td>\n",
       "      <td>0.180052</td>\n",
       "      <td>-0.019792</td>\n",
       "      <td>-0.222489</td>\n",
       "      <td>0.185393</td>\n",
       "      <td>-0.140655</td>\n",
       "      <td>-0.133024</td>\n",
       "      <td>-0.128113</td>\n",
       "      <td>-0.169794</td>\n",
       "      <td>-0.096789</td>\n",
       "      <td>0.006469</td>\n",
       "      <td>0.270412</td>\n",
       "      <td>-0.057915</td>\n",
       "      <td>0.100730</td>\n",
       "      <td>0.038974</td>\n",
       "      <td>-0.015392</td>\n",
       "      <td>0.114684</td>\n",
       "      <td>0.112709</td>\n",
       "      <td>0.157619</td>\n",
       "      <td>-0.027925</td>\n",
       "      <td>-0.094114</td>\n",
       "      <td>0.047447</td>\n",
       "      <td>0.045218</td>\n",
       "      <td>-0.146513</td>\n",
       "      <td>-0.190071</td>\n",
       "      <td>-0.012742</td>\n",
       "      <td>0.013960</td>\n",
       "      <td>-0.154474</td>\n",
       "      <td>-0.178493</td>\n",
       "      <td>0.042555</td>\n",
       "      <td>-0.241204</td>\n",
       "      <td>0.111031</td>\n",
       "      <td>-0.234421</td>\n",
       "      <td>-0.033785</td>\n",
       "      <td>-0.056170</td>\n",
       "      <td>-0.242976</td>\n",
       "      <td>0.064723</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55574</th>\n",
       "      <td>2mut_55574</td>\n",
       "      <td>RGGRLCYCRIRFIVCVGR</td>\n",
       "      <td>0.188402</td>\n",
       "      <td>-0.032151</td>\n",
       "      <td>-0.016142</td>\n",
       "      <td>0.178395</td>\n",
       "      <td>-0.234103</td>\n",
       "      <td>0.137629</td>\n",
       "      <td>-0.121172</td>\n",
       "      <td>0.227992</td>\n",
       "      <td>-0.170820</td>\n",
       "      <td>0.283342</td>\n",
       "      <td>-0.112150</td>\n",
       "      <td>-0.112396</td>\n",
       "      <td>0.169119</td>\n",
       "      <td>0.192328</td>\n",
       "      <td>0.147986</td>\n",
       "      <td>-0.091960</td>\n",
       "      <td>0.039105</td>\n",
       "      <td>-0.073501</td>\n",
       "      <td>0.068120</td>\n",
       "      <td>0.040258</td>\n",
       "      <td>0.171004</td>\n",
       "      <td>0.102717</td>\n",
       "      <td>-0.283967</td>\n",
       "      <td>0.115468</td>\n",
       "      <td>-0.227030</td>\n",
       "      <td>-0.000199</td>\n",
       "      <td>0.175524</td>\n",
       "      <td>-0.078575</td>\n",
       "      <td>0.041848</td>\n",
       "      <td>0.174391</td>\n",
       "      <td>-0.124455</td>\n",
       "      <td>-0.079674</td>\n",
       "      <td>-0.263065</td>\n",
       "      <td>-0.038468</td>\n",
       "      <td>-0.082188</td>\n",
       "      <td>0.150027</td>\n",
       "      <td>0.150465</td>\n",
       "      <td>-0.106896</td>\n",
       "      <td>0.347845</td>\n",
       "      <td>-0.031667</td>\n",
       "      <td>0.227973</td>\n",
       "      <td>-0.287945</td>\n",
       "      <td>0.190300</td>\n",
       "      <td>0.036562</td>\n",
       "      <td>-0.097250</td>\n",
       "      <td>0.360138</td>\n",
       "      <td>0.125830</td>\n",
       "      <td>0.069145</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.236271</td>\n",
       "      <td>-0.162893</td>\n",
       "      <td>0.252559</td>\n",
       "      <td>-0.028423</td>\n",
       "      <td>-0.070498</td>\n",
       "      <td>0.121113</td>\n",
       "      <td>-0.059443</td>\n",
       "      <td>-0.034635</td>\n",
       "      <td>0.237525</td>\n",
       "      <td>0.236959</td>\n",
       "      <td>-0.046912</td>\n",
       "      <td>0.116958</td>\n",
       "      <td>0.080193</td>\n",
       "      <td>0.152792</td>\n",
       "      <td>0.084026</td>\n",
       "      <td>-0.050169</td>\n",
       "      <td>-0.060835</td>\n",
       "      <td>0.193291</td>\n",
       "      <td>-0.063500</td>\n",
       "      <td>0.067474</td>\n",
       "      <td>-0.183866</td>\n",
       "      <td>-0.079865</td>\n",
       "      <td>-0.133193</td>\n",
       "      <td>-0.160934</td>\n",
       "      <td>0.312769</td>\n",
       "      <td>0.027668</td>\n",
       "      <td>0.113067</td>\n",
       "      <td>-0.029518</td>\n",
       "      <td>-0.055950</td>\n",
       "      <td>0.050211</td>\n",
       "      <td>0.117694</td>\n",
       "      <td>0.105392</td>\n",
       "      <td>-0.022154</td>\n",
       "      <td>-0.152707</td>\n",
       "      <td>-0.098150</td>\n",
       "      <td>-0.121952</td>\n",
       "      <td>0.008357</td>\n",
       "      <td>-0.369846</td>\n",
       "      <td>-0.076444</td>\n",
       "      <td>0.069711</td>\n",
       "      <td>-0.105877</td>\n",
       "      <td>-0.114755</td>\n",
       "      <td>0.117399</td>\n",
       "      <td>-0.192672</td>\n",
       "      <td>0.165256</td>\n",
       "      <td>-0.180096</td>\n",
       "      <td>0.054618</td>\n",
       "      <td>-0.080249</td>\n",
       "      <td>-0.221875</td>\n",
       "      <td>0.081274</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55575</th>\n",
       "      <td>2mut_55575</td>\n",
       "      <td>RGGRLCYWRRRFCVCVGP</td>\n",
       "      <td>0.122371</td>\n",
       "      <td>-0.100859</td>\n",
       "      <td>-0.086050</td>\n",
       "      <td>0.295321</td>\n",
       "      <td>-0.244449</td>\n",
       "      <td>0.154902</td>\n",
       "      <td>-0.148233</td>\n",
       "      <td>0.190466</td>\n",
       "      <td>-0.069993</td>\n",
       "      <td>0.284883</td>\n",
       "      <td>-0.192785</td>\n",
       "      <td>-0.113837</td>\n",
       "      <td>0.128719</td>\n",
       "      <td>0.144594</td>\n",
       "      <td>0.107391</td>\n",
       "      <td>-0.183826</td>\n",
       "      <td>-0.042784</td>\n",
       "      <td>-0.114264</td>\n",
       "      <td>0.126728</td>\n",
       "      <td>0.036701</td>\n",
       "      <td>0.098653</td>\n",
       "      <td>0.056515</td>\n",
       "      <td>-0.271583</td>\n",
       "      <td>0.122022</td>\n",
       "      <td>-0.236143</td>\n",
       "      <td>0.023728</td>\n",
       "      <td>0.058662</td>\n",
       "      <td>-0.101638</td>\n",
       "      <td>0.018736</td>\n",
       "      <td>0.137644</td>\n",
       "      <td>-0.127616</td>\n",
       "      <td>-0.068319</td>\n",
       "      <td>-0.310273</td>\n",
       "      <td>-0.019416</td>\n",
       "      <td>-0.073322</td>\n",
       "      <td>0.011458</td>\n",
       "      <td>0.208608</td>\n",
       "      <td>-0.118662</td>\n",
       "      <td>0.336861</td>\n",
       "      <td>0.022826</td>\n",
       "      <td>0.226896</td>\n",
       "      <td>-0.316337</td>\n",
       "      <td>0.120218</td>\n",
       "      <td>-0.020211</td>\n",
       "      <td>-0.051982</td>\n",
       "      <td>0.259963</td>\n",
       "      <td>0.120848</td>\n",
       "      <td>0.064015</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.287452</td>\n",
       "      <td>-0.197988</td>\n",
       "      <td>0.260480</td>\n",
       "      <td>-0.059457</td>\n",
       "      <td>-0.091719</td>\n",
       "      <td>0.145456</td>\n",
       "      <td>-0.054655</td>\n",
       "      <td>-0.041871</td>\n",
       "      <td>0.219950</td>\n",
       "      <td>0.167545</td>\n",
       "      <td>-0.107412</td>\n",
       "      <td>0.054185</td>\n",
       "      <td>0.118780</td>\n",
       "      <td>0.232991</td>\n",
       "      <td>0.102144</td>\n",
       "      <td>-0.017269</td>\n",
       "      <td>-0.215946</td>\n",
       "      <td>0.231809</td>\n",
       "      <td>-0.094971</td>\n",
       "      <td>0.008757</td>\n",
       "      <td>-0.184398</td>\n",
       "      <td>-0.063986</td>\n",
       "      <td>-0.054323</td>\n",
       "      <td>-0.167929</td>\n",
       "      <td>0.300952</td>\n",
       "      <td>0.059726</td>\n",
       "      <td>0.185962</td>\n",
       "      <td>-0.002829</td>\n",
       "      <td>-0.023174</td>\n",
       "      <td>0.101715</td>\n",
       "      <td>0.149682</td>\n",
       "      <td>0.060070</td>\n",
       "      <td>-0.054070</td>\n",
       "      <td>-0.142574</td>\n",
       "      <td>-0.112051</td>\n",
       "      <td>-0.114695</td>\n",
       "      <td>0.038919</td>\n",
       "      <td>-0.259277</td>\n",
       "      <td>-0.036294</td>\n",
       "      <td>0.022650</td>\n",
       "      <td>-0.132063</td>\n",
       "      <td>-0.084254</td>\n",
       "      <td>0.077180</td>\n",
       "      <td>-0.137869</td>\n",
       "      <td>0.118490</td>\n",
       "      <td>-0.146999</td>\n",
       "      <td>0.069662</td>\n",
       "      <td>-0.093890</td>\n",
       "      <td>-0.287631</td>\n",
       "      <td>0.071357</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>55576 rows × 5122 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               ID            Sequence         0         1         2         3  \\\n",
       "0          2mut_0  RGPRLCYCRRRQCVCVGR  0.030471 -0.036276 -0.111070  0.150194   \n",
       "1          2mut_1  YGGPLCYCRRRFCVCVGR  0.078283 -0.004239 -0.083609  0.105327   \n",
       "2          2mut_2  RKGRLCGCRRRFCVCVGR  0.085428 -0.057224 -0.048115  0.131888   \n",
       "3          2mut_3  RGGRLCYCRRWFCVKVGR  0.057880 -0.050223 -0.124206  0.076189   \n",
       "4          2mut_4  RGGRLCYCMRRFCVCDGR  0.212485  0.024579 -0.084416  0.144025   \n",
       "...           ...                 ...       ...       ...       ...       ...   \n",
       "55571  2mut_55571  CGGRLCYCRRRWCVCVGR  0.082458 -0.054124 -0.153812  0.028242   \n",
       "55572  2mut_55572  RGGRLCYCRRRHDVCVGR  0.005155 -0.164447 -0.016448  0.263290   \n",
       "55573  2mut_55573  RGGRMCYCRRRFCVCVAR  0.152141  0.008183 -0.151269  0.136687   \n",
       "55574  2mut_55574  RGGRLCYCRIRFIVCVGR  0.188402 -0.032151 -0.016142  0.178395   \n",
       "55575  2mut_55575  RGGRLCYWRRRFCVCVGP  0.122371 -0.100859 -0.086050  0.295321   \n",
       "\n",
       "              4         5         6         7         8         9        10  \\\n",
       "0     -0.230985  0.181216 -0.178246  0.158280 -0.044290  0.140545 -0.173849   \n",
       "1     -0.233435  0.161460 -0.106038  0.203431 -0.081741  0.155934 -0.148036   \n",
       "2     -0.233556  0.146629 -0.058954  0.145036 -0.012269  0.208369 -0.116377   \n",
       "3     -0.272554  0.140542 -0.079627  0.214544 -0.021029  0.190860 -0.159807   \n",
       "4     -0.232260  0.099251 -0.095035  0.196182 -0.108374  0.186943 -0.141637   \n",
       "...         ...       ...       ...       ...       ...       ...       ...   \n",
       "55571 -0.250708  0.075041 -0.087562  0.206055  0.005263  0.066554 -0.187047   \n",
       "55572 -0.190351  0.191838 -0.132956  0.206665 -0.053722  0.319446 -0.217972   \n",
       "55573 -0.251317  0.134183 -0.108524  0.154696 -0.015930  0.115763 -0.151316   \n",
       "55574 -0.234103  0.137629 -0.121172  0.227992 -0.170820  0.283342 -0.112150   \n",
       "55575 -0.244449  0.154902 -0.148233  0.190466 -0.069993  0.284883 -0.192785   \n",
       "\n",
       "             11        12        13        14        15        16        17  \\\n",
       "0     -0.074231  0.127456  0.139890  0.141800 -0.161217  0.052185 -0.073374   \n",
       "1     -0.138177  0.127338  0.172699  0.143295 -0.190010  0.101302 -0.077433   \n",
       "2     -0.063401  0.179298  0.176179  0.089100 -0.128998  0.192588 -0.062569   \n",
       "3     -0.098083  0.199157  0.179230  0.109095 -0.135120  0.113276 -0.068056   \n",
       "4     -0.176585  0.178890  0.219366  0.157616 -0.134976  0.083950 -0.133882   \n",
       "...         ...       ...       ...       ...       ...       ...       ...   \n",
       "55571 -0.055146  0.113620  0.230673  0.062871 -0.091331  0.158153 -0.072747   \n",
       "55572 -0.144265  0.222120  0.061943  0.056246 -0.118007 -0.077944 -0.173787   \n",
       "55573 -0.068709  0.151451  0.204528  0.095688 -0.143107  0.124312 -0.089717   \n",
       "55574 -0.112396  0.169119  0.192328  0.147986 -0.091960  0.039105 -0.073501   \n",
       "55575 -0.113837  0.128719  0.144594  0.107391 -0.183826 -0.042784 -0.114264   \n",
       "\n",
       "             18        19        20        21        22        23        24  \\\n",
       "0      0.231049  0.004841  0.076786 -0.000564 -0.235507  0.093585 -0.094216   \n",
       "1      0.149300  0.018422  0.155382  0.059158 -0.202118  0.056027 -0.144371   \n",
       "2      0.056391 -0.061444  0.136877  0.051762 -0.240653  0.131735 -0.155009   \n",
       "3      0.066501  0.022015  0.165217  0.009979 -0.291171  0.147817 -0.147829   \n",
       "4      0.113217  0.087540  0.160665  0.063379 -0.183397  0.080644 -0.179947   \n",
       "...         ...       ...       ...       ...       ...       ...       ...   \n",
       "55571  0.209477  0.075546  0.100096 -0.018672 -0.173173  0.077043 -0.020988   \n",
       "55572  0.159124  0.045858  0.058331  0.020416 -0.504820  0.120175 -0.255182   \n",
       "55573  0.142609  0.001460  0.095826 -0.040184 -0.149959 -0.016682 -0.128044   \n",
       "55574  0.068120  0.040258  0.171004  0.102717 -0.283967  0.115468 -0.227030   \n",
       "55575  0.126728  0.036701  0.098653  0.056515 -0.271583  0.122022 -0.236143   \n",
       "\n",
       "             25        26        27        28        29        30        31  \\\n",
       "0     -0.030184  0.090808 -0.017730  0.056953  0.168276 -0.138922 -0.055812   \n",
       "1     -0.061357  0.085607 -0.040719  0.005678  0.153111 -0.134872 -0.038001   \n",
       "2     -0.012141  0.110780 -0.081294  0.068365  0.167697 -0.143250  0.005252   \n",
       "3     -0.041164  0.068331 -0.102698 -0.041732  0.109131 -0.132808 -0.042759   \n",
       "4     -0.104130  0.103943 -0.012196  0.027808  0.159193 -0.135980 -0.071504   \n",
       "...         ...       ...       ...       ...       ...       ...       ...   \n",
       "55571 -0.079208  0.131811  0.014909  0.012486  0.175663 -0.191909  0.008557   \n",
       "55572 -0.000031  0.096224 -0.195531 -0.043021  0.209337 -0.070500  0.015667   \n",
       "55573  0.004092  0.141307 -0.062854  0.089832  0.129261 -0.129841 -0.010646   \n",
       "55574 -0.000199  0.175524 -0.078575  0.041848  0.174391 -0.124455 -0.079674   \n",
       "55575  0.023728  0.058662 -0.101638  0.018736  0.137644 -0.127616 -0.068319   \n",
       "\n",
       "             32        33        34        35        36        37        38  \\\n",
       "0     -0.238720 -0.121912 -0.131041  0.009613  0.166661 -0.117495  0.190636   \n",
       "1     -0.213945 -0.130888 -0.058073  0.013071  0.204674 -0.049626  0.176748   \n",
       "2     -0.351505 -0.103353 -0.154235  0.003228  0.185293 -0.056369  0.242288   \n",
       "3     -0.286052 -0.113242 -0.090135 -0.022518  0.168574 -0.058750  0.236315   \n",
       "4     -0.267629 -0.091737 -0.073513  0.039947  0.202282 -0.122841  0.191903   \n",
       "...         ...       ...       ...       ...       ...       ...       ...   \n",
       "55571 -0.159104 -0.107365 -0.112420  0.019998  0.171493 -0.049025  0.061186   \n",
       "55572 -0.349645 -0.097594 -0.057167  0.002473  0.098192 -0.053422  0.365413   \n",
       "55573 -0.234781 -0.093643 -0.102319 -0.011205  0.151069 -0.078129  0.122194   \n",
       "55574 -0.263065 -0.038468 -0.082188  0.150027  0.150465 -0.106896  0.347845   \n",
       "55575 -0.310273 -0.019416 -0.073322  0.011458  0.208608 -0.118662  0.336861   \n",
       "\n",
       "             39        40        41        42        43        44        45  \\\n",
       "0     -0.058775  0.083950 -0.137385  0.066807  0.088748 -0.054584  0.269736   \n",
       "1     -0.075964  0.177314 -0.164634  0.122157  0.013722 -0.078917  0.326256   \n",
       "2     -0.047226  0.159304 -0.101366  0.098639  0.058434 -0.045043  0.278100   \n",
       "3     -0.028953  0.159713 -0.287824  0.157747  0.060543 -0.057961  0.306598   \n",
       "4     -0.085785  0.163927 -0.266495  0.073727  0.059343 -0.140880  0.346234   \n",
       "...         ...       ...       ...       ...       ...       ...       ...   \n",
       "55571 -0.171707  0.127845 -0.080232  0.065033  0.062051 -0.077829  0.249108   \n",
       "55572  0.010262  0.145131 -0.265709  0.127403  0.157150 -0.059583  0.351660   \n",
       "55573 -0.115878  0.146669 -0.119630  0.065052  0.037127 -0.081287  0.276126   \n",
       "55574 -0.031667  0.227973 -0.287945  0.190300  0.036562 -0.097250  0.360138   \n",
       "55575  0.022826  0.226896 -0.316337  0.120218 -0.020211 -0.051982  0.259963   \n",
       "\n",
       "             46        47  ...      5070      5071      5072      5073  \\\n",
       "0      0.049476  0.108028  ... -0.332250 -0.182212  0.188159 -0.089034   \n",
       "1      0.102039  0.064966  ... -0.208491 -0.181908  0.204974  0.013904   \n",
       "2      0.043613  0.063437  ... -0.328105 -0.162223  0.256969 -0.061340   \n",
       "3      0.056881  0.039657  ... -0.307967 -0.126027  0.237130 -0.060796   \n",
       "4      0.095565  0.073925  ... -0.249055 -0.163885  0.184048 -0.029923   \n",
       "...         ...       ...  ...       ...       ...       ...       ...   \n",
       "55571  0.134323  0.107528  ... -0.235606 -0.205449  0.146447 -0.050081   \n",
       "55572 -0.118701  0.047485  ... -0.184939 -0.108122  0.187322 -0.225827   \n",
       "55573  0.145157  0.107750  ... -0.313060 -0.161855  0.131321 -0.040697   \n",
       "55574  0.125830  0.069145  ... -0.236271 -0.162893  0.252559 -0.028423   \n",
       "55575  0.120848  0.064015  ... -0.287452 -0.197988  0.260480 -0.059457   \n",
       "\n",
       "           5074      5075      5076      5077      5078      5079      5080  \\\n",
       "0     -0.064079  0.156366 -0.090362 -0.010652  0.187834  0.213740 -0.055363   \n",
       "1     -0.081345  0.084125 -0.008529 -0.013594  0.245070  0.272875 -0.054534   \n",
       "2     -0.029240  0.072979 -0.043325 -0.013580  0.229346  0.264277 -0.033285   \n",
       "3     -0.106810  0.076391 -0.032235  0.007105  0.312369  0.172563 -0.045592   \n",
       "4      0.018876  0.082031 -0.042319  0.006107  0.219458  0.257496 -0.044507   \n",
       "...         ...       ...       ...       ...       ...       ...       ...   \n",
       "55571 -0.064856  0.077309 -0.010093  0.033059  0.178510  0.267095 -0.008456   \n",
       "55572  0.000120  0.062208 -0.069744  0.117765  0.194843  0.217680 -0.024724   \n",
       "55573 -0.016497  0.123411  0.000906  0.016701  0.210516  0.298393 -0.057488   \n",
       "55574 -0.070498  0.121113 -0.059443 -0.034635  0.237525  0.236959 -0.046912   \n",
       "55575 -0.091719  0.145456 -0.054655 -0.041871  0.219950  0.167545 -0.107412   \n",
       "\n",
       "           5081      5082      5083      5084      5085      5086      5087  \\\n",
       "0     -0.009788  0.014521  0.185780  0.124452  0.013869 -0.185609  0.156731   \n",
       "1      0.077179  0.028892  0.122911  0.150612 -0.056876 -0.118520  0.162811   \n",
       "2      0.013914 -0.017617  0.160241  0.176237 -0.087623 -0.177227  0.208036   \n",
       "3      0.099534  0.065470  0.134455  0.143874  0.002906 -0.142259  0.190600   \n",
       "4      0.121740  0.041201  0.099270  0.136533 -0.048466 -0.100322  0.195729   \n",
       "...         ...       ...       ...       ...       ...       ...       ...   \n",
       "55571  0.099262  0.064182  0.086947  0.210259 -0.014526 -0.158287  0.158621   \n",
       "55572 -0.040571  0.022824  0.105905 -0.018251 -0.013676 -0.171608  0.225520   \n",
       "55573  0.105326  0.011940  0.097368  0.180052 -0.019792 -0.222489  0.185393   \n",
       "55574  0.116958  0.080193  0.152792  0.084026 -0.050169 -0.060835  0.193291   \n",
       "55575  0.054185  0.118780  0.232991  0.102144 -0.017269 -0.215946  0.231809   \n",
       "\n",
       "           5088      5089      5090      5091      5092      5093      5094  \\\n",
       "0     -0.108033 -0.137903 -0.097508 -0.199132 -0.055758 -0.013001  0.282426   \n",
       "1     -0.106010 -0.062028 -0.084144 -0.140908 -0.094160 -0.074398  0.280061   \n",
       "2     -0.145973  0.000206 -0.188131 -0.057958 -0.108080 -0.064826  0.241150   \n",
       "3     -0.021462 -0.018197 -0.208055 -0.071023 -0.081870 -0.171144  0.213066   \n",
       "4     -0.145436 -0.033395 -0.071495 -0.146533 -0.127662 -0.067469  0.307449   \n",
       "...         ...       ...       ...       ...       ...       ...       ...   \n",
       "55571 -0.075987 -0.103462 -0.080306 -0.153473 -0.079218  0.015159  0.188657   \n",
       "55572 -0.070250  0.049515 -0.161149  0.060027 -0.003002 -0.206381  0.318415   \n",
       "55573 -0.140655 -0.133024 -0.128113 -0.169794 -0.096789  0.006469  0.270412   \n",
       "55574 -0.063500  0.067474 -0.183866 -0.079865 -0.133193 -0.160934  0.312769   \n",
       "55575 -0.094971  0.008757 -0.184398 -0.063986 -0.054323 -0.167929  0.300952   \n",
       "\n",
       "           5095      5096      5097      5098      5099      5100      5101  \\\n",
       "0     -0.031745  0.175234 -0.038487  0.071631  0.056087  0.094258  0.134074   \n",
       "1     -0.020089  0.090262  0.008972 -0.009543  0.066797  0.110991  0.168301   \n",
       "2      0.070687  0.167148 -0.012571 -0.008496  0.036840  0.092076  0.183831   \n",
       "3     -0.009455  0.160588 -0.060975 -0.063147  0.025246  0.081391  0.122101   \n",
       "4     -0.044970  0.040246 -0.006643 -0.103936  0.064969  0.220303  0.141593   \n",
       "...         ...       ...       ...       ...       ...       ...       ...   \n",
       "55571 -0.095693  0.113370 -0.031929  0.017222  0.075270  0.031181  0.160164   \n",
       "55572  0.002611  0.147094 -0.064504  0.063217 -0.062706  0.111827  0.105259   \n",
       "55573 -0.057915  0.100730  0.038974 -0.015392  0.114684  0.112709  0.157619   \n",
       "55574  0.027668  0.113067 -0.029518 -0.055950  0.050211  0.117694  0.105392   \n",
       "55575  0.059726  0.185962 -0.002829 -0.023174  0.101715  0.149682  0.060070   \n",
       "\n",
       "           5102      5103      5104      5105      5106      5107      5108  \\\n",
       "0     -0.014872 -0.083226 -0.015783  0.033664 -0.126464 -0.285006 -0.048229   \n",
       "1     -0.014092 -0.092231  0.006551  0.018966 -0.110038 -0.285004 -0.044679   \n",
       "2     -0.038787 -0.071339 -0.011860 -0.058060 -0.045512 -0.228114 -0.049612   \n",
       "3     -0.058024 -0.072349 -0.008517 -0.108182 -0.060309 -0.290993 -0.001576   \n",
       "4     -0.029952 -0.114162 -0.001476 -0.008912 -0.056513 -0.308030  0.002654   \n",
       "...         ...       ...       ...       ...       ...       ...       ...   \n",
       "55571  0.013887 -0.065848  0.062114  0.035505 -0.174763 -0.289324 -0.014531   \n",
       "55572 -0.136614 -0.100203 -0.238890 -0.075930  0.019882 -0.311911 -0.020147   \n",
       "55573 -0.027925 -0.094114  0.047447  0.045218 -0.146513 -0.190071 -0.012742   \n",
       "55574 -0.022154 -0.152707 -0.098150 -0.121952  0.008357 -0.369846 -0.076444   \n",
       "55575 -0.054070 -0.142574 -0.112051 -0.114695  0.038919 -0.259277 -0.036294   \n",
       "\n",
       "           5109      5110      5111      5112      5113      5114      5115  \\\n",
       "0      0.061758 -0.191082 -0.151428  0.059017 -0.244604  0.117419 -0.155859   \n",
       "1      0.036731 -0.182719 -0.162484  0.037435 -0.229143  0.173071 -0.190800   \n",
       "2     -0.007920 -0.166855 -0.198541  0.027955 -0.190623  0.077585 -0.172339   \n",
       "3      0.017886 -0.120012 -0.084091  0.096398 -0.096984  0.099995 -0.157944   \n",
       "4     -0.009495 -0.175722 -0.094956  0.136943 -0.242696  0.129612 -0.228823   \n",
       "...         ...       ...       ...       ...       ...       ...       ...   \n",
       "55571 -0.020021 -0.158223 -0.175883  0.036970 -0.238702  0.154410 -0.193272   \n",
       "55572  0.041474 -0.139836 -0.138856  0.037153 -0.283946  0.093709 -0.093828   \n",
       "55573  0.013960 -0.154474 -0.178493  0.042555 -0.241204  0.111031 -0.234421   \n",
       "55574  0.069711 -0.105877 -0.114755  0.117399 -0.192672  0.165256 -0.180096   \n",
       "55575  0.022650 -0.132063 -0.084254  0.077180 -0.137869  0.118490 -0.146999   \n",
       "\n",
       "           5116      5117      5118      5119  \n",
       "0     -0.045599 -0.075758 -0.259623  0.045590  \n",
       "1     -0.026766 -0.103359 -0.266636  0.025352  \n",
       "2      0.024809 -0.092293 -0.187360  0.020935  \n",
       "3     -0.051621 -0.029807 -0.291391  0.005705  \n",
       "4     -0.060089 -0.118763 -0.227903  0.101617  \n",
       "...         ...       ...       ...       ...  \n",
       "55571 -0.014171 -0.038470 -0.281103  0.036803  \n",
       "55572  0.197921  0.061686 -0.203528  0.005697  \n",
       "55573 -0.033785 -0.056170 -0.242976  0.064723  \n",
       "55574  0.054618 -0.080249 -0.221875  0.081274  \n",
       "55575  0.069662 -0.093890 -0.287631  0.071357  \n",
       "\n",
       "[55576 rows x 5122 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load embeddings from a pickle file\n",
    "df = pd.read_pickle(f'data/embedding/esm2_15B_pg1_2mut.pkl')\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Prediction of active variants (MBC <= 8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "36458 peptides with probability == 100% to be active\n"
     ]
    }
   ],
   "source": [
    "# creates a df to save predictions\n",
    "features = df.iloc[:,2:]\n",
    "features1 = features[corr_columns_mic]\n",
    "\n",
    "# creates a df to save probabilities\n",
    "res_GNB = df.iloc[:,0:1].copy()\n",
    "\n",
    "# calculates the probability of being active\n",
    "prob= model_GNB_mbc.predict_proba(features1)\n",
    "prob= [(x[1]*100).round(2) for x in prob]\n",
    "\n",
    "res_GNB['MBC_8_prob']=prob\n",
    "\n",
    "#number of peptides with probability == 100%\n",
    "print(f\"{res_GNB.query('MBC_8_prob == 100').shape[0]} peptides with probability == 100% to be active\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Predicting selectivity score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "301 peptides with selectivity score < 0.5\n"
     ]
    }
   ],
   "source": [
    "# creates a df to selectivity score\n",
    "res_sel = df.iloc[:,0:1].copy()\n",
    "\n",
    "# calculates the the selectivity score\n",
    "prob= model_SVR_sel.predict(features[corr_columns_sel])\n",
    "res_sel['selectivity_pred']= prob.round(2) \n",
    "\n",
    "#number of peptides with selectivity score < 0.5\n",
    "print(f\"{res_sel.query('selectivity_pred < 0.5').shape[0]} peptides with selectivity score < 0.5\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Prediction nom Hemolitic variants (hemolysis% < 2%)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4792 peptides with predicted hemolysis% < 2%\n"
     ]
    }
   ],
   "source": [
    "# creates a df to save hemolysis score\n",
    "res_hemo = df.iloc[:,0:1].copy()\n",
    "\n",
    "# calculates the the hemolysis score\n",
    "pred= model_SVR_hemo.predict(features[corr_columns_hemo])\n",
    "res_hemo['hemolysis_pred']= pred.round(2)\n",
    "\n",
    "#number of peptides with hemolysis < 2%\n",
    "print(f\"{res_hemo.query('hemolysis_pred <= 2').shape[0]} peptides with predicted hemolysis% < 2%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Results of double mutants"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>Sequence</th>\n",
       "      <th>MBC_8_prob</th>\n",
       "      <th>selectivity_pred</th>\n",
       "      <th>hemolysis_pred</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2mut_0</td>\n",
       "      <td>RGPRLCYCRRRQCVCVGR</td>\n",
       "      <td>99.17</td>\n",
       "      <td>1.64</td>\n",
       "      <td>14.900000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2mut_1</td>\n",
       "      <td>YGGPLCYCRRRFCVCVGR</td>\n",
       "      <td>100.00</td>\n",
       "      <td>2.38</td>\n",
       "      <td>30.129999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2mut_2</td>\n",
       "      <td>RKGRLCGCRRRFCVCVGR</td>\n",
       "      <td>100.00</td>\n",
       "      <td>2.21</td>\n",
       "      <td>6.730000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2mut_3</td>\n",
       "      <td>RGGRLCYCRRWFCVKVGR</td>\n",
       "      <td>98.46</td>\n",
       "      <td>2.42</td>\n",
       "      <td>17.590000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2mut_4</td>\n",
       "      <td>RGGRLCYCMRRFCVCDGR</td>\n",
       "      <td>100.00</td>\n",
       "      <td>1.95</td>\n",
       "      <td>23.129999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55571</th>\n",
       "      <td>2mut_55571</td>\n",
       "      <td>CGGRLCYCRRRWCVCVGR</td>\n",
       "      <td>0.00</td>\n",
       "      <td>2.92</td>\n",
       "      <td>28.740000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55572</th>\n",
       "      <td>2mut_55572</td>\n",
       "      <td>RGGRLCYCRRRHDVCVGR</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.98</td>\n",
       "      <td>-0.270000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55573</th>\n",
       "      <td>2mut_55573</td>\n",
       "      <td>RGGRMCYCRRRFCVCVAR</td>\n",
       "      <td>100.00</td>\n",
       "      <td>2.01</td>\n",
       "      <td>17.260000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55574</th>\n",
       "      <td>2mut_55574</td>\n",
       "      <td>RGGRLCYCRIRFIVCVGR</td>\n",
       "      <td>100.00</td>\n",
       "      <td>3.01</td>\n",
       "      <td>18.600000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55575</th>\n",
       "      <td>2mut_55575</td>\n",
       "      <td>RGGRLCYWRRRFCVCVGP</td>\n",
       "      <td>100.00</td>\n",
       "      <td>3.00</td>\n",
       "      <td>25.660000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>55576 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               ID            Sequence  MBC_8_prob  selectivity_pred  \\\n",
       "0          2mut_0  RGPRLCYCRRRQCVCVGR       99.17              1.64   \n",
       "1          2mut_1  YGGPLCYCRRRFCVCVGR      100.00              2.38   \n",
       "2          2mut_2  RKGRLCGCRRRFCVCVGR      100.00              2.21   \n",
       "3          2mut_3  RGGRLCYCRRWFCVKVGR       98.46              2.42   \n",
       "4          2mut_4  RGGRLCYCMRRFCVCDGR      100.00              1.95   \n",
       "...           ...                 ...         ...               ...   \n",
       "55571  2mut_55571  CGGRLCYCRRRWCVCVGR        0.00              2.92   \n",
       "55572  2mut_55572  RGGRLCYCRRRHDVCVGR        0.00              0.98   \n",
       "55573  2mut_55573  RGGRMCYCRRRFCVCVAR      100.00              2.01   \n",
       "55574  2mut_55574  RGGRLCYCRIRFIVCVGR      100.00              3.01   \n",
       "55575  2mut_55575  RGGRLCYWRRRFCVCVGP      100.00              3.00   \n",
       "\n",
       "       hemolysis_pred  \n",
       "0           14.900000  \n",
       "1           30.129999  \n",
       "2            6.730000  \n",
       "3           17.590000  \n",
       "4           23.129999  \n",
       "...               ...  \n",
       "55571       28.740000  \n",
       "55572       -0.270000  \n",
       "55573       17.260000  \n",
       "55574       18.600000  \n",
       "55575       25.660000  \n",
       "\n",
       "[55576 rows x 5 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results = res_GNB.merge(res_sel[['ID', 'selectivity_pred']], how='inner', left_on='ID', right_on='ID')\n",
    "results = results.merge(res_hemo[['ID', 'hemolysis_pred']], how='inner', left_on='ID', right_on='ID')\n",
    "results = meta_data.merge(results, how='inner', left_on='ID', right_on='ID')\n",
    "results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Define the sets as lists of elements\n",
    "set1 = set(results.query('MBC_8_prob == 100')['ID'].values)\n",
    "set2 = set(results.query('selectivity_pred < 0.5')['ID'].values)\n",
    "set3 = set(results.query('hemolysis_pred <= 2')['ID'].values)\n",
    "\n",
    "# Create the Venn diagram\n",
    "venn3_unweighted([set1, set2, set3], set_labels=('Activity', 'Selectivity', 'Hemolysis'), set_colors=('b', 'g', 'r'))\n",
    "#plt.savefig('MBC_HEMO_SEL_2mut_venn3_predictions_v03.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# saving results\n",
    "results.to_csv('results/predictions_MIC_SEL_HEMO_models_PG1_2mut_v01.2.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Predictions on triple mutant Protegrin-1 Variants"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "### Those functions will be used to predict the triple mutants in a loop\n",
    "def run_GNB(features):    \n",
    "    # creates a df to save probabilities\n",
    "    res_GNB = pd.DataFrame()\n",
    "    \n",
    "    features1 = features[corr_columns_mic]\n",
    "\n",
    "    # calculates the probability of being active\n",
    "    prob= model_GNB_mbc.predict_proba(features1)\n",
    "    prob= [(x[1]*100).round(2) for x in prob]\n",
    "    \n",
    "    res_GNB['ID']=features.index\n",
    "    res_GNB['MBC_8_prob']=prob\n",
    "\n",
    "    return res_GNB"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "### Predicting selectivity score\n",
    "def run_SVR_sel(features, corr_columns):\n",
    "    # creates a df to selectivity score\n",
    "    res_sel = pd.DataFrame()\n",
    "\n",
    "    # calculates the the selectivity score\n",
    "    prob= model_SVR_sel.predict(features[corr_columns])\n",
    "\n",
    "    res_sel['ID']=features.index\n",
    "    res_sel['selectivity_pred']= prob.round(2) \n",
    "\n",
    "    return res_sel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "### Prediction nom Hemolitic variants (hemolysis% < 2%)\n",
    "def run_SVR_hemo(features, corr_columns):\n",
    "    # creates a df to save hemolysis score\n",
    "    res_hemo = pd.DataFrame()\n",
    "\n",
    "    # calculates the the hemolysis score\n",
    "    pred= model_SVR_hemo.predict(features[corr_columns])\n",
    "\n",
    "    res_hemo['ID']=features.index\n",
    "    res_hemo['hemolysis_pred']= pred.round(2)\n",
    "\n",
    "    return res_hemo"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Preprocess the embedding from a tensor file to a lighter pickle file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import multiprocessing\n",
    "\n",
    "# # this was required to load the embeddings with multiple CPUs due the large file sizes\n",
    "# for i in range(1,7):\n",
    "#     '''this loop will load the embeddings with multiple CPUs and save them in a pickle file'''\n",
    "\n",
    "#     # Set the directory containing the Torch files\n",
    "#     data_dir = f'data/embedding/esm2_15B_pg1_3mut/esm2_15B_pg1_3mut_{i}mil/'\n",
    "\n",
    "#     # Define the function to load a Torch file\n",
    "#     def load_torch_file(filename):\n",
    "#         filepath = os.path.join(data_dir, filename)\n",
    "#         label = filename.split('.pt')[0]\n",
    "#         embeddings = {}\n",
    "#         embeddings[label] = np.array(torch.load(filepath)['mean_representations'][48])\n",
    "#         return embeddings\n",
    "\n",
    "#     # Get a list of all the Torch files in the directory\n",
    "#     file_list = os.listdir(data_dir)\n",
    "\n",
    "#     # Define the number of processes to use\n",
    "#     num_processes = 50\n",
    "\n",
    "#     # Create a multiprocessing pool with the number of processes defined above\n",
    "#     pool = multiprocessing.Pool(processes=num_processes)\n",
    "\n",
    "#     # Use the pool to load all the Torch files in parallel\n",
    "#     results = pool.map(load_torch_file, file_list)\n",
    "\n",
    "#     # Close the pool\n",
    "#     pool.close()\n",
    "\n",
    "#     # Concatenate the results into a single list\n",
    "#     all_data = {k: v for d in results for k, v in d.items()}\n",
    "\n",
    "#     # Convert the dictionary to a NumPy array\n",
    "#     arr = np.array(list(all_data.values()))\n",
    "\n",
    "#     # Set the keys as the index\n",
    "#     index = list(all_data.keys())\n",
    "\n",
    "#     # Create a DataFrame from the array\n",
    "#     df = pd.DataFrame(arr, index=index)\n",
    "\n",
    "#     # Print the DataFrame and renamed it\n",
    "#     df.to_pickle(f'data/embedding/esm2_15B_pg1_3mut_{i}mil.pkl')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Loaning the pre processced embedding from a pickle file and run prediction on each splited file\n",
    "I have to split the files in 6 parts due its huge size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processing file esm2_15B_pg1_3mut_1mil, size: (1000000, 5120)\n",
      "Prediction size: 1000000\n",
      "Processing file esm2_15B_pg1_3mut_2mil, size: (1000000, 5120)\n",
      "Prediction size: 2000000\n",
      "Processing file esm2_15B_pg1_3mut_3mil, size: (1000000, 5120)\n",
      "Prediction size: 3000000\n",
      "Processing file esm2_15B_pg1_3mut_4mil, size: (1000000, 5120)\n",
      "Prediction size: 4000000\n",
      "Processing file esm2_15B_pg1_3mut_5mil, size: (1000000, 5120)\n",
      "Prediction size: 5000000\n",
      "Processing file esm2_15B_pg1_3mut_6mil, size: (652520, 5120)\n",
      "Prediction size: 5652520\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>MBC_8_prob</th>\n",
       "      <th>selectivity_pred</th>\n",
       "      <th>hemolysis_pred</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>PG1_3mut_602975</td>\n",
       "      <td>100.00</td>\n",
       "      <td>1.81</td>\n",
       "      <td>8.350000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>PG1_3mut_185031</td>\n",
       "      <td>100.00</td>\n",
       "      <td>2.73</td>\n",
       "      <td>12.650000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>PG1_3mut_332050</td>\n",
       "      <td>100.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.160000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>PG1_3mut_938612</td>\n",
       "      <td>0.00</td>\n",
       "      <td>2.70</td>\n",
       "      <td>35.689999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>PG1_3mut_894404</td>\n",
       "      <td>100.00</td>\n",
       "      <td>2.46</td>\n",
       "      <td>13.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5652515</th>\n",
       "      <td>PG1_3mut_5154857</td>\n",
       "      <td>21.08</td>\n",
       "      <td>2.10</td>\n",
       "      <td>11.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5652516</th>\n",
       "      <td>PG1_3mut_5261581</td>\n",
       "      <td>100.00</td>\n",
       "      <td>2.82</td>\n",
       "      <td>15.290000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5652517</th>\n",
       "      <td>PG1_3mut_5406149</td>\n",
       "      <td>100.00</td>\n",
       "      <td>2.69</td>\n",
       "      <td>19.129999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5652518</th>\n",
       "      <td>PG1_3mut_5061684</td>\n",
       "      <td>100.00</td>\n",
       "      <td>2.64</td>\n",
       "      <td>14.870000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5652519</th>\n",
       "      <td>PG1_3mut_5468652</td>\n",
       "      <td>0.00</td>\n",
       "      <td>2.24</td>\n",
       "      <td>20.799999</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5652520 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                       ID  MBC_8_prob  selectivity_pred  hemolysis_pred\n",
       "0         PG1_3mut_602975      100.00              1.81        8.350000\n",
       "1         PG1_3mut_185031      100.00              2.73       12.650000\n",
       "2         PG1_3mut_332050      100.00              1.00        1.160000\n",
       "3         PG1_3mut_938612        0.00              2.70       35.689999\n",
       "4         PG1_3mut_894404      100.00              2.46       13.000000\n",
       "...                   ...         ...               ...             ...\n",
       "5652515  PG1_3mut_5154857       21.08              2.10       11.000000\n",
       "5652516  PG1_3mut_5261581      100.00              2.82       15.290000\n",
       "5652517  PG1_3mut_5406149      100.00              2.69       19.129999\n",
       "5652518  PG1_3mut_5061684      100.00              2.64       14.870000\n",
       "5652519  PG1_3mut_5468652        0.00              2.24       20.799999\n",
       "\n",
       "[5652520 rows x 4 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# read the CSV files and concatenate them row-wise\n",
    "predictions = pd.DataFrame()\n",
    "\n",
    "for i in range(1, 7):\n",
    "    file_path = f'data/embedding/esm2_15B_pg1_3mut_{i}mil.pkl'\n",
    "    features = pd.read_pickle(file_path)\n",
    "    print(f'Processing file esm2_15B_pg1_3mut_{i}mil, size: {features.shape}')\n",
    "    \n",
    "    res_GNB = run_GNB(features)\n",
    "    res_sel = run_SVR_sel(features, corr_columns_sel)\n",
    "    res_hemo = run_SVR_hemo(features, corr_columns_hemo)\n",
    "    \n",
    "    res = res_GNB.merge(res_sel[['ID', 'selectivity_pred']], how='inner', left_on='ID', right_on='ID')\n",
    "    res = res.merge(res_hemo[['ID', 'hemolysis_pred']], how='inner', left_on='ID', right_on='ID')\n",
    "\n",
    "    predictions = pd.concat([predictions, res], axis=0).reset_index(drop=True)\n",
    "    print(f'Prediction size: {predictions.shape[0]}')\n",
    "\n",
    "predictions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load meta data and merge with results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>Sequence</th>\n",
       "      <th>MBC_8_prob</th>\n",
       "      <th>selectivity_pred</th>\n",
       "      <th>hemolysis_pred</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>PG1_3mut_0</td>\n",
       "      <td>HGGRLCYCRRRFCVCQTR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>0.68</td>\n",
       "      <td>17.540001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>PG1_3mut_1</td>\n",
       "      <td>RGGQLCYVRRRFCVAVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.88</td>\n",
       "      <td>-4.060000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>PG1_3mut_2</td>\n",
       "      <td>RCYRLCRCRRRFCVCVGR</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.33</td>\n",
       "      <td>17.820000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>PG1_3mut_3</td>\n",
       "      <td>RGSRLCYCRRRFCMCVIR</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.41</td>\n",
       "      <td>43.610001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>PG1_3mut_4</td>\n",
       "      <td>RGGRRCYCWRAFCVCVGR</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.99</td>\n",
       "      <td>39.060001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5652515</th>\n",
       "      <td>PG1_3mut_5652515</td>\n",
       "      <td>RGIRLCYCERRSCVCVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2.08</td>\n",
       "      <td>18.200001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5652516</th>\n",
       "      <td>PG1_3mut_5652516</td>\n",
       "      <td>RGGRLCACRRLFCVCVGV</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.23</td>\n",
       "      <td>17.799999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5652517</th>\n",
       "      <td>PG1_3mut_5652517</td>\n",
       "      <td>RGGDLCLCLRRFCVCVGR</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.41</td>\n",
       "      <td>22.990000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5652518</th>\n",
       "      <td>PG1_3mut_5652518</td>\n",
       "      <td>FGGRLCYCRRRFCQGVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.47</td>\n",
       "      <td>4.330000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5652519</th>\n",
       "      <td>PG1_3mut_5652519</td>\n",
       "      <td>RGGRLCYCDRRFCVNCGR</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.10</td>\n",
       "      <td>1.230000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5652520 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                       ID            Sequence  MBC_8_prob  selectivity_pred  \\\n",
       "0              PG1_3mut_0  HGGRLCYCRRRFCVCQTR       100.0              0.68   \n",
       "1              PG1_3mut_1  RGGQLCYVRRRFCVAVGR       100.0              1.88   \n",
       "2              PG1_3mut_2  RCYRLCRCRRRFCVCVGR         0.0              2.33   \n",
       "3              PG1_3mut_3  RGSRLCYCRRRFCMCVIR         0.0              2.41   \n",
       "4              PG1_3mut_4  RGGRRCYCWRAFCVCVGR         0.0              2.99   \n",
       "...                   ...                 ...         ...               ...   \n",
       "5652515  PG1_3mut_5652515  RGIRLCYCERRSCVCVGR       100.0              2.08   \n",
       "5652516  PG1_3mut_5652516  RGGRLCACRRLFCVCVGV         0.0              3.23   \n",
       "5652517  PG1_3mut_5652517  RGGDLCLCLRRFCVCVGR         0.0              3.41   \n",
       "5652518  PG1_3mut_5652518  FGGRLCYCRRRFCQGVGR       100.0              1.47   \n",
       "5652519  PG1_3mut_5652519  RGGRLCYCDRRFCVNCGR         0.0              1.10   \n",
       "\n",
       "         hemolysis_pred  \n",
       "0             17.540001  \n",
       "1             -4.060000  \n",
       "2             17.820000  \n",
       "3             43.610001  \n",
       "4             39.060001  \n",
       "...                 ...  \n",
       "5652515       18.200001  \n",
       "5652516       17.799999  \n",
       "5652517       22.990000  \n",
       "5652518        4.330000  \n",
       "5652519        1.230000  \n",
       "\n",
       "[5652520 rows x 5 columns]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "meta_data = pd.read_csv('data/protegrin_variants_3mut.csv', index_col=0)\n",
    "results = meta_data.merge(predictions, how='inner', left_on='ID', right_on='ID')\n",
    "results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "# saving results\n",
    "results.to_csv('results/predictions_MIC_SEL_HEMO_models_PG1_3mut_v01.2.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Define the sets as lists of elements\n",
    "set1 = set(results.query('MBC_8_prob == 100')['ID'].values)\n",
    "set2 = set(results.query('selectivity_pred < 0.5')['ID'].values)\n",
    "set3 = set(results.query('hemolysis_pred <= 2')['ID'].values)\n",
    "\n",
    "# Create the Venn diagram\n",
    "venn3_unweighted([set1, set2, set3], set_labels=('Activity', 'Selectivity', 'Hemolysis'), set_colors=('b', 'g', 'r'))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Merge predictions and visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>Sequence</th>\n",
       "      <th>MBC_8_prob</th>\n",
       "      <th>selectivity_pred</th>\n",
       "      <th>hemolysis_pred</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1mut_0</td>\n",
       "      <td>RGGRLCYCRRRFCVLVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.94</td>\n",
       "      <td>8.91</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1mut_1</td>\n",
       "      <td>RMGRLCYCRRRFCVCVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2.34</td>\n",
       "      <td>35.30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1mut_2</td>\n",
       "      <td>RGGRLCYCRRRFCVCVFR</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.81</td>\n",
       "      <td>42.71</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1mut_3</td>\n",
       "      <td>EGGRLCYCRRRFCVCVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2.30</td>\n",
       "      <td>30.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1mut_4</td>\n",
       "      <td>RGGNLCYCRRRFCVCVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2.23</td>\n",
       "      <td>28.78</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5652515</th>\n",
       "      <td>PG1_3mut_5652515</td>\n",
       "      <td>RGIRLCYCERRSCVCVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2.08</td>\n",
       "      <td>18.20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5652516</th>\n",
       "      <td>PG1_3mut_5652516</td>\n",
       "      <td>RGGRLCACRRLFCVCVGV</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.23</td>\n",
       "      <td>17.80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5652517</th>\n",
       "      <td>PG1_3mut_5652517</td>\n",
       "      <td>RGGDLCLCLRRFCVCVGR</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.41</td>\n",
       "      <td>22.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5652518</th>\n",
       "      <td>PG1_3mut_5652518</td>\n",
       "      <td>FGGRLCYCRRRFCQGVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.47</td>\n",
       "      <td>4.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5652519</th>\n",
       "      <td>PG1_3mut_5652519</td>\n",
       "      <td>RGGRLCYCDRRFCVNCGR</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.10</td>\n",
       "      <td>1.23</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5708439 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                       ID            Sequence  MBC_8_prob  selectivity_pred  \\\n",
       "0                  1mut_0  RGGRLCYCRRRFCVLVGR       100.0              1.94   \n",
       "1                  1mut_1  RMGRLCYCRRRFCVCVGR       100.0              2.34   \n",
       "2                  1mut_2  RGGRLCYCRRRFCVCVFR         0.0              2.81   \n",
       "3                  1mut_3  EGGRLCYCRRRFCVCVGR       100.0              2.30   \n",
       "4                  1mut_4  RGGNLCYCRRRFCVCVGR       100.0              2.23   \n",
       "...                   ...                 ...         ...               ...   \n",
       "5652515  PG1_3mut_5652515  RGIRLCYCERRSCVCVGR       100.0              2.08   \n",
       "5652516  PG1_3mut_5652516  RGGRLCACRRLFCVCVGV         0.0              3.23   \n",
       "5652517  PG1_3mut_5652517  RGGDLCLCLRRFCVCVGR         0.0              3.41   \n",
       "5652518  PG1_3mut_5652518  FGGRLCYCRRRFCQGVGR       100.0              1.47   \n",
       "5652519  PG1_3mut_5652519  RGGRLCYCDRRFCVNCGR         0.0              1.10   \n",
       "\n",
       "         hemolysis_pred  \n",
       "0                  8.91  \n",
       "1                 35.30  \n",
       "2                 42.71  \n",
       "3                 30.99  \n",
       "4                 28.78  \n",
       "...                 ...  \n",
       "5652515           18.20  \n",
       "5652516           17.80  \n",
       "5652517           22.99  \n",
       "5652518            4.33  \n",
       "5652519            1.23  \n",
       "\n",
       "[5708439 rows x 5 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# loop trhrough the 3 predictions files and concatenate them\n",
    "files = []\n",
    "for i in range(1, 4):\n",
    "    file = pd.read_csv(f'results/predictions_MIC_SEL_HEMO_models_PG1_{i}mut_v01.2.csv', index_col=0)\n",
    "    files.append(file)\n",
    "\n",
    "data = pd.concat(files, axis=0)\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "## save the file compressed\n",
    "data.to_csv('results/predictions_MIC_SEL_HEMO_models_PG1_1-3mut_v01.2.csv.gz', index=False, compression='gzip')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualizing Predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>Sequence</th>\n",
       "      <th>MBC_8_prob</th>\n",
       "      <th>selectivity_pred</th>\n",
       "      <th>hemolysis_pred</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1mut_0</td>\n",
       "      <td>RGGRLCYCRRRFCVLVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.94</td>\n",
       "      <td>8.91</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1mut_1</td>\n",
       "      <td>RMGRLCYCRRRFCVCVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2.34</td>\n",
       "      <td>35.30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1mut_2</td>\n",
       "      <td>RGGRLCYCRRRFCVCVFR</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.81</td>\n",
       "      <td>42.71</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1mut_3</td>\n",
       "      <td>EGGRLCYCRRRFCVCVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2.30</td>\n",
       "      <td>30.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1mut_4</td>\n",
       "      <td>RGGNLCYCRRRFCVCVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2.23</td>\n",
       "      <td>28.78</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5708434</th>\n",
       "      <td>PG1_3mut_5652515</td>\n",
       "      <td>RGIRLCYCERRSCVCVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2.08</td>\n",
       "      <td>18.20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5708435</th>\n",
       "      <td>PG1_3mut_5652516</td>\n",
       "      <td>RGGRLCACRRLFCVCVGV</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.23</td>\n",
       "      <td>17.80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5708436</th>\n",
       "      <td>PG1_3mut_5652517</td>\n",
       "      <td>RGGDLCLCLRRFCVCVGR</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.41</td>\n",
       "      <td>22.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5708437</th>\n",
       "      <td>PG1_3mut_5652518</td>\n",
       "      <td>FGGRLCYCRRRFCQGVGR</td>\n",
       "      <td>100.0</td>\n",
       "      <td>1.47</td>\n",
       "      <td>4.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5708438</th>\n",
       "      <td>PG1_3mut_5652519</td>\n",
       "      <td>RGGRLCYCDRRFCVNCGR</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.10</td>\n",
       "      <td>1.23</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5708439 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                       ID            Sequence  MBC_8_prob  selectivity_pred  \\\n",
       "0                  1mut_0  RGGRLCYCRRRFCVLVGR       100.0              1.94   \n",
       "1                  1mut_1  RMGRLCYCRRRFCVCVGR       100.0              2.34   \n",
       "2                  1mut_2  RGGRLCYCRRRFCVCVFR         0.0              2.81   \n",
       "3                  1mut_3  EGGRLCYCRRRFCVCVGR       100.0              2.30   \n",
       "4                  1mut_4  RGGNLCYCRRRFCVCVGR       100.0              2.23   \n",
       "...                   ...                 ...         ...               ...   \n",
       "5708434  PG1_3mut_5652515  RGIRLCYCERRSCVCVGR       100.0              2.08   \n",
       "5708435  PG1_3mut_5652516  RGGRLCACRRLFCVCVGV         0.0              3.23   \n",
       "5708436  PG1_3mut_5652517  RGGDLCLCLRRFCVCVGR         0.0              3.41   \n",
       "5708437  PG1_3mut_5652518  FGGRLCYCRRRFCQGVGR       100.0              1.47   \n",
       "5708438  PG1_3mut_5652519  RGGRLCYCDRRFCVNCGR         0.0              1.10   \n",
       "\n",
       "         hemolysis_pred  \n",
       "0                  8.91  \n",
       "1                 35.30  \n",
       "2                 42.71  \n",
       "3                 30.99  \n",
       "4                 28.78  \n",
       "...                 ...  \n",
       "5708434           18.20  \n",
       "5708435           17.80  \n",
       "5708436           22.99  \n",
       "5708437            4.33  \n",
       "5708438            1.23  \n",
       "\n",
       "[5708439 rows x 5 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## lOAD the file\n",
    "data = pd.read_csv('results/predictions_MIC_SEL_HEMO_models_PG1_1-3mut_v01.2.csv.gz', compression='gzip')\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjIAAAI2CAYAAABKXWJPAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/OQEPoAAAACXBIWXMAAA9hAAAPYQGoP6dpAACQjElEQVR4nOzdd5xU9fX/8de908v2vizswtKbIE1FQSxYMPauUSyRaDRqfmrsoNFo1K8l9kQjRrFEo8beG2JXsIDSe93eps/c3x8jI8MusMuWe+fOeeaxkZ25c+/ZgZ15z6cqmqZpCCGEEEKkIFXvAoQQQgghdpcEGSGEEEKkLAkyQgghhEhZEmSEEEIIkbIkyAghhBAiZUmQEUIIIUTKkiAjhBBCiJQlQUYIIYQQKUuCjBBCCCFSlgQZIdLcqlWrUBSF2bNnd/ixs2fPRlEUVq1albht//33Z//99++y+oxu+vTpVFRUJN3W3NzMueeeS3FxMYqicMkll3TqeRZC7JgEGSHaaeub9rZfhYWFTJkyhTfeeKNbr/3AAw+k7Rtgc3MzM2fOZPjw4Xg8HvLy8hg1ahQXX3wxGzZs0Lu8Nv31r39l9uzZnH/++TzxxBP89re/1bskIUzLqncBQqSaG2+8kb59+6JpGps3b2b27NkcfvjhvPLKKxxxxBHdcs0HHniA/Px8pk+f3uXnLi8vx+/3Y7PZuvzcnRUOh5k0aRI///wzZ555JhdddBHNzc0sXLiQp556imOOOYbS0lJda/znP/9JLBZLuu39999nr732YubMmYnbNE0z7PMsRCqTICNEBx122GGMHTs28f0555xDUVERTz/9dLcFme4QiUSIxWLY7XacTqfe5bTppZdeYv78+cyZM4dTTz016b5AIEAoFNKpsl+1FUy2bNnC0KFDk25TFMWwz7MQqUy6loTopOzsbFwuF1Zr8ueCO+64g3322Ye8vDxcLhdjxozh+eefb/McTz75JOPHj8ftdpOTk8OkSZN4++23AaioqGDhwoV89NFHiS6tbceg1NfXc8kll9C7d28cDgf9+/fnb3/7W1IrwdbxGXfccQd33303lZWVOBwOFi1a1ObYje+//57p06fTr18/nE4nxcXFnH322dTU1HTouWlubsbj8XDxxRe3um/dunVYLBZuueWWHT5++fLlAEycOLHVfU6nk8zMzMT306dPx+v1smLFCg455BA8Hg+lpaXceOONaJqW9NhYLMbdd9/NsGHDcDqdFBUVMWPGDOrq6lpd54033mDy5MlkZGSQmZnJuHHjeOqpp5Kuu3WMzIcffoiiKKxcuZLXXnst8fe1atWqHY6R+fnnnznxxBMpKCjA5XIxaNAgrrnmmh0+J0KIZNIiI0QHNTQ0UF1djaZpbNmyhXvvvZfm5mZOP/30pOPuuecejjzySE477TRCoRDPPPMMJ5xwAq+++irTpk1LHHfDDTcwa9Ys9tlnH2688UbsdjtffPEF77//PlOnTuXuu+/moosuwuv1Jt7gioqKAPD5fEyePJn169czY8YM+vTpw6effspVV13Fxo0bufvuu5NqeuyxxwgEApx33nk4HA5yc3NbdYsAvPPOO6xYsYKzzjqL4uJiFi5cyD/+8Q8WLlzI559/jqIo7XquvF4vxxxzDM8++yx33nknFoslcd/TTz+NpmmcdtppO3x8eXk5AP/+97+59tprd3ndaDTKoYceyl577cVtt93Gm2++ycyZM4lEItx4442J42bMmMHs2bM566yz+OMf/8jKlSu57777mD9/PvPmzUu0ssyePZuzzz6bYcOGcdVVV5Gdnc38+fN58803W7UQAQwZMoQnnniCSy+9lLKyMv7f//t/ABQUFFBVVdXq+O+//5799tsPm83GeeedR0VFBcuXL+eVV17h5ptv3unPKoT4hSaEaJfHHntMA1p9ORwObfbs2a2O9/l8Sd+HQiFt+PDh2gEHHJC4benSpZqqqtoxxxyjRaPRpONjsVjiz8OGDdMmT57c6hp/+ctfNI/Hoy1ZsiTp9iuvvFKzWCzamjVrNE3TtJUrV2qAlpmZqW3ZsiXp2K33PfbYYzusXdM07emnn9YA7eOPP271nKxcuTJx2+TJk5NqfeuttzRAe+ONN5LON3LkyDZ/pm35fD5t0KBBGqCVl5dr06dP1x599FFt8+bNrY4988wzNUC76KKLErfFYjFt2rRpmt1u16qqqjRN07S5c+dqgDZnzpykx7/55ptJt9fX12sZGRnahAkTNL/fn3Tstn83Z555plZeXp50f3l5uTZt2rSk29p6nidNmqRlZGRoq1ev3uH5hRA7J11LQnTQ/fffzzvvvMM777zDk08+yZQpUzj33HN54YUXko5zuVyJP9fV1dHQ0MB+++3Ht99+m7j9pZdeIhaLcf3116Oqyb+O7Wn1eO6559hvv/3Iycmhuro68XXQQQcRjUb5+OOPk44/7rjjKCgo2OV5t609EAhQXV3NXnvtBZBUf3scdNBBlJaWMmfOnMRtP/74I99//32rVqy26vjiiy+4/PLLgXgLyTnnnENJSQkXXXQRwWCw1WMuvPDCxJ8VReHCCy8kFArx7rvvAvHnLCsri4MPPjjpORszZgxer5cPPvgAiLdKNTU1ceWVV7Ya29LeFqmdqaqq4uOPP+bss8+mT58+XX5+IdKFdC0J0UHjx49PGux7yimnMHr0aC688EKOOOII7HY7AK+++io33XQTCxYsSHrD3fZNavny5aiq2mpgaHstXbqU77//fofhZMuWLUnf9+3bt13nra2t5YYbbuCZZ55pdY6GhoYO1aiqKqeddhoPPvggPp8Pt9vNnDlzcDqdnHDCCbt8fFZWFrfddhu33XYbq1ev5r333uOOO+7gvvvuIysri5tuuinpWv369Ut6/MCBAwESa90sXbqUhoYGCgsL27ze1p936/ic4cOHd+jnba8VK1Z06/mFSBcSZIToJFVVmTJlCvfccw9Lly5l2LBhzJ07lyOPPJJJkybxwAMPUFJSgs1m47HHHksaKNpZsViMgw8+mCuuuKLN+7e+iW+1bUvLzpx44ol8+umnXH755YwaNQqv10ssFuPQQw9tc0zNrpxxxhncfvvtvPTSS5xyyik89dRTHHHEEWRlZXXoPOXl5Zx99tkcc8wx9OvXjzlz5iQFmfaIxWIUFhYmtRBtqz0tVkII45AgI0QXiEQiQHyWDsB///tfnE4nb731Fg6HI3HcY489lvS4yspKYrEYixYtYtSoUTs8/466GiorK2lubuaggw7q5E/wq7q6Ot577z1uuOEGrr/++sTtS5cu3e1zDh8+nNGjRzNnzhzKyspYs2YN9957726fLycnh8rKSn788cek22OxGCtWrEgKcEuWLAFIzCyqrKzk3XffZeLEiTsNdpWVlUC8G6x///67XeuObG052v5nEEJ0jIyREaKTwuEwb7/9Nna7nSFDhgBgsVhQFIVoNJo4btWqVbz00ktJjz366KNRVZUbb7yxVUuHts2UYY/HQ319fatrn3jiiXz22We89dZbre6rr69PBKyO2DqzSNtuyvL2M6A66re//S1vv/02d999N3l5eRx22GG7fMx3331HdXV1q9tXr17NokWLGDRoUKv77rvvvsSfNU3jvvvuw2azceCBBwLx5ywajfKXv/yl1WMjkUjieZ46dSoZGRnccsstBAKBpOO2f252R0FBAZMmTeJf//oXa9as6fLzC5EupEVGiA564403+Pnnn4H4eIqnnnqKpUuXcuWVVybWNZk2bRp33nknhx56KKeeeipbtmzh/vvvp3///nz//feJc/Xv359rrrmGv/zlL+y3334ce+yxOBwOvvrqK0pLSxNrrIwZM4YHH3yQm266if79+1NYWMgBBxzA5Zdfzssvv8wRRxzB9OnTGTNmDC0tLfzwww88//zzrFq1ivz8/A79fJmZmUyaNInbbruNcDhMr169ePvtt1m5cmWnnrdTTz2VK664ghdffJHzzz+/XSvcvvPOO8ycOZMjjzySvfbaK7FOzL/+9S+CwSCzZs1KOt7pdPLmm29y5plnMmHCBN544w1ee+01rr766kSX0eTJk5kxYwa33HILCxYsYOrUqdhsNpYuXcpzzz3HPffcw/HHH09mZiZ33XUX5557LuPGjePUU08lJyeH7777Dp/Px+OPP96p5wPg73//O/vuuy977rkn5513Hn379mXVqlW89tprLFiwoNPnFyIt6DpnSogU0tb0a6fTqY0aNUp78MEHW02ZffTRR7UBAwZoDodDGzx4sPbYY49pM2fO1Nr6tfvXv/6ljR49WnM4HFpOTo42efJk7Z133kncv2nTJm3atGlaRkaGBiRNW25qatKuuuoqrX///prdbtfy8/O1ffbZR7vjjju0UCikadqvU39vv/32Vtdua1rwunXrtGOOOUbLzs7WsrKytBNOOEHbsGGDBmgzZ85s9ZzsbPr1tg4//HAN0D799NOdPNO/WrFihXb99ddre+21l1ZYWKhZrVatoKBAmzZtmvb+++8nHXvmmWdqHo9HW758uTZ16lTN7XZrRUVF2syZM1tNbdc0TfvHP/6hjRkzRnO5XFpGRoY2YsQI7YorrtA2bNiQdNzLL7+s7bPPPprL5dIyMzO18ePHa08//XTSdXd3+rWmadqPP/6YeK6dTqc2aNAg7brrrmvX8yOE0DRF06QNUwjRM4455hh++OEHli1b1uXnnj59Os8//3xinJIQIj3IGBkhRI/YuHEjr732muwELYToUjJGRgjRrVauXMm8efN45JFHsNlszJgxQ++ShBAmIi0yQohu9dFHH/Hb3/6WlStX8vjjj1NcXKx3SUIIE5ExMkIIIYRIWdIiI4QQQoiUJUFGCCGEEClLgowQQgghUpYEGSGEEEKkLAkyQgghhEhZEmSEEEIIkbIkyAghhBAiZUmQEUIIIUTKkiAjhBBCiJQlQUYIIYQQKUuCjBBCCCFSlgQZIYQQQqQsCTJCCCGESFkSZIQQQgiRsiTICCGEECJlSZARQgghRMqSICOEEEKIlCVBRgghhBApS4KMEEIIIVKWBBkhhBBCpCwJMkIIIYRIWRJkhBBCCJGyJMgIIYQQImVJkBFCCCFEypIgI4QQQoiUJUFGCCGEEClLgowQQgghUpYEGSGEEEKkLAkyOps+fToVFRUdftyqVatQFIXZs2d3eU1CCJGOdvf1uLM683quV81GIkGmHR544AEURWHChAm79fgNGzYwa9YsFixY0LWFbef1119n1qxZ3XoNIYQwkh9++IHjjz+e8vJynE4nvXr14uCDD+bee+/Vu7RWnnrqKe6+++5uvYbP52PWrFl8+OGH3XodI1E0TdP0LsLoJk6cyIYNG1i1ahVLly6lf//+HXr8119/zbhx43jssceYPn160n3hcJhYLIbD4ejQOTVNIxgMYrPZsFgsAFx44YXcf//9yF+pECIdfPrpp0yZMoU+ffpw5plnUlxczNq1a/n8889Zvnw5y5Yt69D5pk+fzocffsiqVau6pd4jjjiCH3/8sdX523o9b6/t30Oqq6spKChg5syZafPB1qp3AUa3cuVKPv30U1544QVmzJjBnDlzmDlzZped32az7dbjFEXB6XR2WR1CCJFqbr75ZrKysvjqq6/Izs5Oum/Lli36FLUbOvN6vrvvIWYiXUu7MGfOHHJycpg2bRrHH388c+bMaXVMfX09l156KRUVFTgcDsrKyjjjjDOorq7mww8/ZNy4cQCcddZZKIqS1Be6bf9mOBwmNzeXs846q9U1GhsbcTqdXHbZZUDrPtXp06dz//33AySuoSgKmqZRUVHBUUcd1eqcgUCArKwsZsyY0dmnSQghetzy5csZNmxYqxADUFhYmPT9k08+yZgxY3C5XOTm5nLyySezdu3aXV4jFotx9913M2zYMJxOJ0VFRcyYMYO6urpWx77xxhtMnjyZjIwMMjMzGTduHE899RQA+++/P6+99hqrV69OvD5vfe3f/vX8jjvuQFEUVq9e3eoaV111FXa7PXH9bd9DVq1aRUFBAQA33HBD4jqzZs3iscceQ1EU5s+f3+qcf/3rX7FYLKxfv36Xz4cRSZDZhTlz5nDsscdit9s55ZRTWLp0KV999VXi/ubmZvbbbz/uvfdepk6dyj333MPvf/97fv75Z9atW8eQIUO48cYbATjvvPN44okneOKJJ5g0aVKra9lsNo455hheeuklQqFQ0n0vvfQSwWCQk08+uc06Z8yYwcEHHwyQuMYTTzyBoiicfvrpvPHGG9TW1iY95pVXXqGxsZHTTz+9U8+REELooby8nG+++YYff/xxp8fdfPPNnHHGGQwYMIA777yTSy65hPfee49JkyZRX1+/08fOmDGDyy+/nIkTJ3LPPfdw1llnMWfOHA455BDC4XDiuNmzZzNt2jRqa2u56qqruPXWWxk1ahRvvvkmANdccw2jRo0iPz8/8fq8o/EyJ554Ioqi8J///KfVff/5z3+YOnUqOTk5re4rKCjgwQcfBOCYY45JXOfYY4/l+OOPx+VytflhfM6cOey///706tVrp8+FYWlih77++msN0N555x1N0zQtFotpZWVl2sUXX5w45vrrr9cA7YUXXmj1+Fgspmmapn311VcaoD322GOtjjnzzDO18vLyxPdvvfWWBmivvPJK0nGHH3641q9fv8T3K1eubHXOP/zhD1pbf6WLFy/WAO3BBx9Muv3II4/UKioqEnUKIUQqefvttzWLxaJZLBZt77331q644grtrbfe0kKhUOKYVatWaRaLRbv55puTHvvDDz9oVqs16fbtX4/nzp2rAdqcOXOSHvvmm28m3V5fX69lZGRoEyZM0Px+f9Kx276+Tps2Len8W7X1er733ntrY8aMSTruyy+/1ADt3//+9w5rrqqq0gBt5syZra5zyimnaKWlpVo0Gk3c9u233+7w/SlVSIvMTsyZM4eioiKmTJkCxLtsTjrpJJ555hmi0SgA//3vf9ljjz045phjWj1eUZQOX/OAAw4gPz+fZ599NnFbXV0d77zzDieddNJu/RwDBw5kwoQJSUm8traWN954g9NOO2236hRCCL0dfPDBfPbZZxx55JF899133HbbbRxyyCH06tWLl19+GYAXXniBWCzGiSeeSHV1deKruLiYAQMG8MEHH+zw/M899xxZWVkcfPDBSY8dM2YMXq838dh33nmHpqYmrrzyylZjXXb39fWkk07im2++Yfny5Ynbnn32WRwOR5tDBdrjjDPOYMOGDUk/85w5c3C5XBx33HG7dU4jkCCzA9FolGeeeYYpU6awcuVKli1bxrJly5gwYQKbN2/mvffeA+J9tMOHD++y61qtVo477jj+97//EQwGgfgvYjgc3u0gA/F/wPPmzUv0uT733HOEw2F++9vfdkndQgihh3HjxvHCCy9QV1fHl19+yVVXXUVTUxPHH388ixYtYunSpWiaxoABAygoKEj6+umnn3Y6KHjp0qU0NDRQWFjY6rHNzc2Jx24NG135XnDCCSegqmriQ62maTz33HMcdthhZGZm7tY5Dz74YEpKShIfamOxGE8//TRHHXUUGRkZXVZ7T5NZSzvw/vvvs3HjRp555hmeeeaZVvfPmTOHqVOndsu1Tz75ZB5++GHeeOMNjj76aP7zn/8wePBg9thjj06d89JLL2XOnDlcffXVPPnkk4wdO5ZBgwZ1YeVCCKEPu93OuHHjGDduHAMHDuSss87iueeeIxaLoSgKb7zxRptTm71e7w7PGYvFKCwsbHNcCZAYWNsdSktL2W+//fjPf/7D1Vdfzeeff86aNWv429/+ttvntFgsnHrqqfzzn//kgQceYN68eWzYsCHlx0lKkNmBOXPmUFhYmJgJtK0XXniBF198kYceeojKyspdDjTraNPipEmTKCkp4dlnn2Xffffl/fff55prrtnl43Z2ndzcXKZNm8acOXM47bTTmDdvXrcvzCSEEHoYO3YsABs3bqSyshJN0+jbty8DBw7s0HkqKyt59913mThxIi6Xa6fHAfz44487XWeso+8FJ510EhdccAGLFy/m2Wefxe1285vf/Ganj9nVNc444wz+7//+j1deeYU33niDgoICDjnkkA7VZTTStdQGv9/PCy+8wBFHHMHxxx/f6uvCCy+kqamJl19+meOOO47vvvuOF198sdV5tF8WpvN4PAC7HB2/laqqHH/88bzyyis88cQTRCKRdnUr7eo6v/3tb1m0aBGXX345FotlhzOghBAiFXzwwQdtLgD6+uuvAzBo0CCOPfZYLBYLN9xwQ6tjNU2jpqZmh+c/8cQTiUaj/OUvf2l1XyQSSbzWTp06lYyMDG655RYCgUCra2zl8XhoaGho98933HHHYbFYePrpp3nuuec44ogjEq/zO+J2u4Edvw+MHDmSkSNH8sgjj/Df//6Xk08+Gas1tds0Urv6bvLyyy/T1NTEkUce2eb9e+21FwUFBcyZM4ennnqK559/nhNOOIGzzz6bMWPGUFtby8svv8xDDz3EHnvsQWVlJdnZ2Tz00ENkZGTg8XiYMGECffv23WENJ510Evfeey8zZ85kxIgRDBkyZJd1jxkzBoA//vGPHHLIIa3CyrRp08jLy0v0s26/zoIQQqSSiy66CJ/PxzHHHMPgwYMJhUJ8+umnPPvss1RUVHDWWWeRnZ3NTTfdxFVXXcWqVas4+uijycjIYOXKlbz44oucd955ifW5tjd58mRmzJjBLbfcwoIFC5g6dSo2m42lS5fy3HPPcc8993D88ceTmZnJXXfdxbnnnsu4ceM49dRTycnJ4bvvvsPn8/H4448D8dfoZ599lj/96U+MGzcOr9e70xaWwsJCpkyZwp133klTU1O7PtC6XC6GDh3Ks88+y8CBA8nNzWX48OFJ43fOOOOMxM+c6t1KIEGmTXPmzMHpdCbWZdmeqqqJbppgMMjcuXOZOXMmL774Io8//jiFhYUceOCBlJWVAfH1YR5//HGuuuoqfv/73xOJRHjsscd2GmT22Wcfevfuzdq1a9s9yPfYY4/loosu4plnnuHJJ59E07SkIGO32znppJN44IEHZJDvLmgaNDeD3w+RCITD8f9u+9XWbdveB2Cztf6y28Fqjf+3rfttNnA6QZX2UsMJRUOEoiHC0TDhWDjpz9v/NxQNEYlFEsdEtShW1brDL5tqa32bJX6b3WLHa/dit9j1fgoM5Y477uC5557j9ddf5x//+AehUIg+ffpwwQUXcO211yYWyrvyyisZOHAgd911FzfccAMAvXv3ZurUqTv8wLrVQw89xJgxY3j44Ye5+uqrsVqtVFRUcPrppzNx4sTEceeccw6FhYXceuut/OUvf8FmszF48GAuvfTSxDEXXHABCxYs4LHHHuOuu+6ivLx8l11FJ510Eu+++y4ZGRkcfvjh7XpeHnnkES666CIuvfRSQqEQM2fOTAoyp512Gn/+85+prKxk/Pjx7TqnkcleS2nm0ksv5dFHH2XTpk2JJsh0FQpBY2P8q6kp+b/NzRCL6VebooDXC5mZkJWV/N/MTOjgdiyiA/xhPw3BBhqDjTQEGhJ/bgw2EoqGdn2CbuSwOMh0ZJLhyCDTkRn/sz3+Z6/dK0spiHaprq6mpKSE66+/nuuuu07vcjpNWmTSSCAQ4Mknn+S4445LmxATCEBNTXJI2frnX2a3G5KmxWtsaoLtVw1XFPB4fg032wcdCTm75gv7aAj8Ela2CS2NwUbCsfCuT6CTYDRIla+KKl9Vq/tURcVr9yaCzdawk+3MJseZIyFHJMyePZtoNGqalnkJMmlgy5YtvPvuuzz//PPU1NRw8cUX611St6mthc2bf/3qwLi6lLG126u5GTZsaH1/djYUF//6tZtLTphGJBZhS8sWNjVvYmPTRra0bDF0WNldMS2WaDla35Scfm2qjQJPAcXeYoo8RRR5i6SbKg29//77LFq0iJtvvpmjjz46sUdTqpOupTTw4YcfMmXKFAoLC7nuuuu48MIL9S6pS4TDsGVLcnAJ6dvyb0hud3KwycuLt+qYVSASYFPzpsRXta+amKZjP6FB5ThzKPIWJcJNljNL75JEN9t///359NNPmThxIk8++WTq7q20HQkyImU0NSWHlpqaeOuE6Bi7HQoL46GmpCT+51TujmoKNrGxeWMiuNQH6vUuKSU5rc5Ea02xt5gCdwEWNYX/YYi0IUFGGFY0Gh8fsmoVrF0LLS16V2ROqgoFBfFgU14e/6+RBSIBVtWvYn3jejY1b6IlLP8wuoOqqBR5iijPLqciu4JMR5r3UQrDkiAjDCUYhNWr419r18anMoue5fFA375QWQlFRXpXE7c1vCyvXc6Gpg1oyMtWT8tx5lCRXUF5djmFHlmDShiHBBmhO78fVqyAlSth0yZ9pz2LZF7vr6Gmp9dPDEaCrKxfyYq6FWxo2iDjXAzEbXNTkV1Bv5x+lHhLZEaU0JUEGaGLYDAeXJYvj8+8kX+Fxuf1Qr9+8a/uCjXBSDDe8lK3XMJLinDb3PTL6Uf/3P7SUiN0IUFG9JhwOD7eZflyWLdOWl5SWUbGry01nd0AOBQNJbqN1jetl/CSwjLsGVTmVtI/tz+5rly9yxFpQoKM6HaNjfDjj7BkiUyPNqOMDBgwAIYMiY+vaQ9N01jTsIafqn9iXeM6CS8mVOQpYnjhcPrm9EVVZL8N0X0kyIhus359PMCsWSNdR+lAVaGiAoYNi0/rbkswEmRxzWIWbllIU6ipR+sT+vDYPAwtGMqQgiE4rU69yxEmJEFGdKlIBJYujQeYujq9qxF6ycuLB5r+/eMbZNb6a/lxy48sq11GJCZT0dKRRbHQP7c/I4pGSLeT6FISZESXaG6GhQvh55+NvYeR6EkamaV+MnpvYFP4C6KKrPci4kq8JYwoGkF5VrnMeBKdJkFGdMqmTfDDD/FBvPIvSQAoqoazoJmIt4kwv+xppIDLXUfEtowwtfoWKAwjw57BsMJhDM4fLHs/id0mQUZ0WDQan3n0449QXa13NcIoVGsMZ1ETQWcTUaI7PM7pbEFzLCeobOzB6oSRWVUrA/MGMrxwONnObL3LESlGgozokKVL4auv4l1JQgAoagxncRNBVyMx2j/7yOlsIeZYTEip6sbqRCpRUOif259xvcbhtXv1LkekCAkyol3WrYMvvohv1CgEAIqGq6iZkKdhpy0wu+J0NRK1/0xYkdHhIs6iWBhWOIw9S/aULiexSxJkxE7V1MQDzLp1elcijEPDVdhCOKOBCF03A8npriNi+4mIItOyRZzD4mB0yWiGFQyTnbjFDkmQEW1qbo53IS1dqnclwkic+S1Esxp+HcTb1RRwuWsI2RYRxdc91xApx2v3Mq50HAPyBuhdijAgCTIiSTAI8+fHp1JHd7+3QJiMI8ePlltPiJ5ZmllRNVzuLQSsi4j10DWF8eW785nQawK9MnvpXYowEAkyAoiHloUL4yFG1oERW1kcUWwltQQs+rSOWCxR7N4V+JUVulxfGFNZZhkTek0gz52ndynCACTICJmJJNqg4S5pIuBu6NBMpO7icPqIOX8gTL3epQiDkBlOYisJMmmsrg4+/BCqZPar2IYjK4iWX9tj3UjtpoDbsxm/9Ue0LhxkLFKbRbGwZ8mejCoeJasEpykJMmlI0+C77+Cbb2QcjPiVaotiL6knYDN205zVGsHqWUpAWaN3KcJACtwF7F+xPzmuHL1LET1Mgkyaqa+HDz6QVhiRzFXUTMhb36n1YHqa09VExP6DTNcWCRbFwpjSMexRtIe0zqQRCTJpQlphRFusrjCWkhqCSmqO8FZUDZdnLT7LT3qXIgyk0FPI/hX7y3YHaUKCTBqor4+PhdmyRe9KhJG4ipoJeusMMZi3s5zOFsKO+bLDtkiwKBbGlo5lZNFIaZ0xOQkyJqZp8Z2pv/pKWmHEr1RbFHtZLQHVXAvOqWoMR8ZS/MoqvUsRBlLkKWL/iv3JcmbpXYroJhJkTKqhId4Ks3mz3pUII3Hm+gnn1KTUWJiOcnnqCNoWyEJ6IsGqWhlbOpYRhSOkdcaEJMiYzNZWmK+/hojMUBW/UNQYzrJ6/Lb0GBhrtUaweBcSZJPepQgDKfYWM7l8srTOmIwEGRNpbob334dN8tottmHPCKEVVnff/kgG5vJuwm/9AUwwDkh0DatqZUKvCQwrHKZ3KaKLSJAxiY0b4Z13IBDQuxJhJO7SRnyueiB9f83t9iAx1zcyTVskGZg3kP367Ce7apuABBkTWLQIPv0UYvKhU/xCUWM4+9Tit8gsHogPBLZlLCSobNC7FGEghZ5CplZOxW1z612K6AQJMiksFoN58+AnWUJDbMPqiqCWVhlviwEDcGeuxacu0rsMYSBum5uplVMp9BTqXYrYTRJkUpTfH+9KkvEwYluO7ACRvGpTz0rqLJe7noD9G9mvSSRYFAv79tmXQfmD9C5F7AYJMimouhreegtapNdAbMNd0ojPXad3GSnBbg8Sc39NBGPvKyV61vDC4exVtheqoupdiugACTIpZtky+PhjmVotfqWoGs4+NTIepoMslihW70KCyka9SxEGUppRykH9DsJpdepdimgnCTIpQtPgiy/g++/1rkQYidUZQe0l42F2mwIu72r8lp/1rkQYSIY9g0P6H0KuK1fvUkQ7SJBJAaEQvPsurFundyXCSOwZIaKFW2Q8TBdweWrwW78BRV4ORZxVtbJ/xf70y+mndyliFyTIGFxdXXw8TGOj3pUII3FkBwjnVZliw0ejcLoaCdq/RFMkGIpf7VmyJ2NLx+pdhtgJCTIGtmULvP56vEVGiK2c+T6CWdVoabzIXXdxOlsIOr5AU9JvFWSxY/1z+zOlYors02RQEmQMatMmeOMNCMvrqdiGq6gZv7dG7zJMzW4PEHF/QQxZJlv8ql9OPw7oe4DMaDIgCTIGtGEDvPmmzEwSydylDb9sNyC6m80WQvN8JdOzRZKK7AoO6neQhBmDkSBjMOvWwdtvS4gRydy96/DZZaBUT7JaIijebwkrsjaP+FWfrD4c3O9g2aPJQCTIGMiaNfHVeqMy1lBspWi4+tTgt8oaMXpQ1RjWzO8IsUXvUoSBlGWWMbVyKlbVqncpAgkyhrFqVXyKtWz8KBIUDWdFFQHVr3claU1RNewZP8qGkyJJaUYph/Y/VMKMAUiQMYAVK+D99yXEiG0oGs6KagKqT+9KBL+EmczvCSKbm4lfFXuLOaz/YdgsNr1LSWsyYklny5bBe+9JiBHb0nCV10iIMRAtphBuHIFdK9C7FGEgm5o38frS1wlFZY0MPUmLjI6WLIGPPopvPyDEVq7yGvxWmS1jRKoaw5L5DWFq9S5FGEiBu4DDBxyOw+rQu5S0JEFGJz/9BHPn6l2FMBp3n1p8tia9yxA7YbFEUTO+Jky93qUIA8lz5TFt4DTZbFIH0rWkAwkxoi3usnoJMSkgGrWgNY3BqmXoXYowkBp/Da8ueZVgJKh3KWlHgkwPW7MGPvlE7yqE0bhLG/A5GvQuQ7RTJGoF3zgsmkfvUoSB1PpreXv528Q0GfTYkyTI9KCamvjAXunME9tylzTKir0pKBK2ofrGoyJdCeJXG5s38tGqj/QuI61IkOkhPl982wHZO0lsy1XYjM8tK8emqnDYjtW3F4om02/Fr5bWLuXbjd/qXUbakCDTAyKReIhpkcVZxTYcWQH8GTL7JdWFQg4coXGgyc7I4ldfb/iaZbXL9C4jLUiQ6WaaFu9Oqq7WuxJhJFZXhEh+NSD9jGYQ8GfgiuyhdxnCYD5a9RGbmzfrXYbpSZDpZp9/DqtX612FMBLFEkMp3UIU2VTLTPwtRbhiA/QuQxhIVIvy1vK3aAzKhq/dSYJMN1q0CH74Qe8qhLFoOPpUE0YGS5mRv6kfDq1E7zKEgQQiAd5c9qas/tuNJMh0k7VrYd48vasQRuPuXS+bQJqZBpHmYVjJ1LsSYSD1gXqZlt2NJMh0g9ra+E7WMs1abMtV1IzPLk3MZheNWlBaxqBi17sUYSAbmjYwd7WshNodJMh0MZlmLdriyA7g98oMpXQRDtuxB2Qmk0i2uGYxCzYt0LsM05Eg04UiEXjrLWiW/f7ENqyuMOG8KmSGUnoJBLy4o6P1LkMYzJfrv2RF3Qq9yzAVCTJdaN48qKrSuwphJIqqoZZWE0P6xtORr7kAl1ahdxnCYD5c9SH1gXq9yzANCTJdZMUKWLxY7yqE0TjL6gghsxXSWbBpgAz+FUkisQjvr3xfBv92EQkyXaClRXazFq258lvwy27WaS8WU1F9e6JoFr1LEQZS7avm6w1f612GKUiQ6QIffghB2bldbMPqChPMksG9Ii4UcuCMjNK7DGEw3236jo1NG/UuI+VJkOmk77+H9ev1rkIYiqKhlsi4GJHM35KPUyvXuwxhIBoaH6z6QBbL6yQJMp1QUwNffql3FcJo3GX1hBR5YRKthZsHYMGtdxnCQJpDzbK+TCdJkNlN0Si8/z7E5EO32IYjOyCL3okdikYt2AJ76l2GMJjldctZWrNU7zJSlgSZ3fTFF1BXp3cVwkhUW5RonmxzLnYuEPDgjg3VuwxhMPPWzqMpKJMDdocEmd2wdi38+KPeVQijcfSqJSI7Wot28DX1xq7l6V2GMJBQNMQHqz5Ak71tOkyCTAcFAvDRR3pXIYzGmefDb/HpXYZIFRrgHy5bGIgkm5o3MX/TfL3LSDkSZDroo4/i+ykJsZViiRHJlqnWomNCISfu2GC9yxAG8+3Gb9nSskXvMlKKBJkO+OknWL1a7yqE0Th71UuXktgt/pbeWPHqXYYwkJgW44OVHxCJRfQuJWVIkGmnxkb47DO9qxBGY88Myuq9YrdpMQVrcA+9yxAG0xBs4PN1n+tdRsqQINNOn34a391aiARFg4IavasQKS7g9+KK9dW7DGEwP1X9RFWL7ELcHhJk2mH1alizRu8qhNG4SxsJEda7DGECoZZKVM2hdxnCQDQ0PlnzicxiagcJMrsQjcZbY4TYls0dxu9s0LsMYRLRqAVHeKTeZQiDqfJVsbhmsd5lGJ4EmV1YsACaZAiE2I5aXIuGfFISXcfvy8WhlepdhjCYL9d/STAiuxLvjFXvAoyssTEeZITYlquoGb8S0LuMbrPk+y94+z//YM3SH2io2cL5NzzMqImHJO6ffdv/47O3/5v0mKFjJ3Hxrf9uda5wKMitFx3NuuU/ce1Dr9G7/zAAFi/4jPf++ygrF39HwNdMYa8Kpp44gwkHHp30+G8+eo3/zf4/ajato7BXX4793ZWMmDCl639og4j5BqO4N6MpMgtOxAUiAb5c/yX7le+ndymGJUFmJz77LN61JMRWihoj5K3Xu4xuFQr4KOs3hImHnsBDs37f5jHDxk3mzMtvT3xvtbU9vuOFf95Cdl4R65b/lHT78kXf0qvfEA45+Xwyc/L5/vP3eOxvf8LlyWDkXgfGj1n4DY/c/EeOPucKRu51IF++/z8enHke1zz4Kr36Duqin9ZYwmEbrthA/Jafdn2wSBs/V//MkIIh5Lvz9S7FkCTI7MC6dbJmjGjNWdKI3+RrxgwfP4Xh43fe6mG12cnKLdzpMT9++QGLvpnLjJkP8eOXHybdd/ipf0j6/sBjz2bR13OZP/fNRJB574V/MWzcZA45aQYAR531//jpm7l8+L/HOe2Sv3bwp0odwZYy1MzlxJAd1EWchsa8NfM4avBRepdiSDJGpg2aBp/LFH6xHYs9StApA6YAlnz3OZcdP4brpx/AnLuvobkheQfVxroqnrjzKs76813YHc52ndPf0oQnMzvx/YpF8xm858SkY4aOm8SKRd92un4ji8VUnBHZVFIk29yymRV1K/Quw5AkyLRh8WKolRXnxXbsxQ3EiOldhu6GjZvMWX++k0tvm8Oxv/szS7//gnuvnk7sl35YTdOYfdtlTDriNCoGtW8mztcfvsrqJd+zzyEnJG5rrKsiMye5KT0zu4CGWvPvMO5vKZQVf0UrX67/kpgmr0Hbk66l7UQi8PXXelchjMbmDuO3NetdhiGMm3Jk4s+9+g2mV98hXHvGJBZ/9zlD9pzIBy/NJuBr4bBTLmjX+RYv+JTH77ic0y+9hdKKgd1VdkrRNAVraCgR+5d6lyIMpDHYyMItCxlRNELvUgxFgsx2FiyQTSFFa5aiesIy3bpNBaV98GblUrVhFUP2nMjP8z9lxU/f8ofDkkPJXy84kvEHHsVZf74zcduS7z7n/mvP5YTfX8feU49LOj4zp4DGuuTWl8b6KrJy02PAY8CXg92WR0iR1aPFr77d+C0D8wbisMoCiltJkNmGzwfff693FcJoHFkBAqqk2x2pq9pIS2NdYvDvyX+YxVFnXZa4v6FmM/dceQa/u/Y++g4Zlbh98YLPuP/aczjmd1cy6YhTW52339DR/Dz/Uw467pzEbT998wn9hu7ZfT+MwajBIeD8RO8yhIEEo0G+3fgte/feW+9SDEOCzDa+/lr2UxJtyKvXu4IeFfC3ULV+VeL76o1rWbtsIZ6MbNyZ2bz673vYc79DycwtoGrDGl745y0UlFYwdOwkAHKLeiWdz+FyA/GWm5yCEiDenXTftedwwDFnsed+h9JQuwUAq9WeGPB74LFnc8efTuKd5/7JiAlT+OqDV1i95AdOv/SWbn4GjCMQ8OB09CKgrNe7FGEgC6sWMqxwGJmOTL1LMQQJMr9oboYlS/SuQhiNM99HQEmvVTVXL/6eOy87JfH9cw/dBMDeU4/j1ItvZv2Kn/j8nf/ia24kO6+QIWMmcdRZf8Jmb39T92dv/5dQwM+bTz/Am08/kLh94MgJ/L87nwWgctgYzr36Hv732P/x0r9up7BXBeff8A/TriGzIzH/AHBLkBG/imkxvtv0nSyS9wtFkx2pgPjidz/8oHcVwmhslRsIy8aQQmeurMX4lVV6lyEMxKJYOHXEqbhsLr1L0Z1MvwZCIfj5Z72rEEbjzPNJiBGGEPVX6F2CMJioFmVh1UK9yzAECTLAwoUQlvcrsR0tq1HvEoQAIBRyyIaSopWFWxYSicnAzrQPMtEo/Pij3lUIo3FkBQim2dgYYWxKsJ/eJQiDCUaD/Fwt3QlpH2SWLAG/X+8qhOHkSGuMMJZAwIOd9FhDR7Tf95u/T/vVftM6yGiarBsjWrN5QgQtkm6F8VhC/fUuQRhMc6g57fdgSusgs3o1NDToXYUwGmu+tMYIY/L7srAia4eIZN9t+k7vEnSV1kFmwQK9KxBGY3VG8FtlFV9hXLaw7EclktX4a1jXuE7vMnSTtkFm0ybYskXvKoTR2AuaQPZUEgYW8OVi0WTtEJEsnVtl0jbISGuM2J5qixKwyw7Xwtg0TcERlVYZkWx903qqfdW7PtCE0jLI1NXBmjV6VyGMxpHfQoz0Hv0vUkPQX0CavnyLnUjXVpm03Gvpu276u37jjVuYP/8FNm36GbvdRb9++3DssX+juDh5b5jlyz/jf/+7hpUrv0BVLZSVjeLii9/Cbk9uLg6Hg9x66wTWrfuOa6+dT+/eowB45ZVZvPrqDa2ub7e7uffeFgA+/XQ2jz9+VtL9VquD++8PJL7XNI1XXpnJ3Ln/xO+vp7JyIqee+iBFRQMSx2zevIT//vdyli2bRzQaolevkRx11F8YNGhKp54rI4q4pTVGpIZo1IJL6yPbFgBLvl3C20+8zZqf1tBQ3cD5d5zPqP1HJe5/5eFX+Ortr6jbXIfVZqXPkD4cfcHR9B3et9W5wqEwt06/lXVL1nHtnGvpPag3ANUbqrnmyGtaHf/nx/5MvxHGWd9nRd0KxgfHk+HI0LuUHpV2Qcbvh2XLuufcS5Z8xP77/4GKinFEoxFeeulq7rlnKrNmLcLh8ADxEPP3vx/KYYddxckn34uqWlm37jsUpfWnqxdeuILs7FLWrUtOXgcffBmTJv0+6ba77jqQiopxSbc5nZnceOPixPeKoiTd/9Zbt/H++39n+vTHyc/vy8svX8ff/34Is2YtwmZzAnDffUdQWDiAP/3pfWw2F++9dzf33XcEN920nKys4t1/sgzGnhkkJNsRiBSiBXuBc5XeZegu5A9RNqCMiUdO5KHLH2p1f1F5EadccQr5vfIJB8O8+9S73P2Hu7nppZvIyEl+w3/h7y+QnZ/NuiVtD5y95IFLKO336wrL3mxv1/4wnaSh8cOWH9in9z56l9Kj0q5tcvlyiHVT78HFF7/JPvtMp7R0GL1778H06bOprV3D6tXfJI557rlLOeCAP3LooVdSWjqM4uJBjB17IjZb8s7BP/74BosWvc1xx93R6jpOp5esrOLEV2PjZjZuXMTEieckHacoStJxmZlFifs0TeO99+7m8MOvZdSooygrG8lZZ/2b+voNLFjwEgDNzdVs2bKUQw+9krKykRQVDeDYY28lFPKxYYO5lkO25EhrjEgtgYAXK8Z6I9XD8InDOfqCoxk9ZXSb948/dDxDJgyhoKyA0spSTrj0BAItAdYtTQ4rP877kUWfL+K4S47b4bW8WV6y8rMSXxarpUt/lq6wvHZ52i2Ql3YtMt3VGtMWvz++SI3HkwtAY+MWVq78gvHjT+Nvf9uHqqrlFBcP5uijb6Z//30Tj2ts3MwTT/yO889/CbvdvcvrfPLJIxQVDWTAgOQt3YPBZq66qhxNi9Gnz54cffRfKS0dBkB19UoaGzcxZMhBieNdriz69p3AihWfMW7cyXg8eRQVDeLzz/9Nnz57YrU6+Pjjh8nIKKRPnzGdfn6MQlFjBGXKtUhB9mhfIpYf9C4jZUTCEea+OBeX10Xvgb0TtzfWNPLEzU9w/h3nY3fad/j4+/90P5FQhMI+hRxyxiHsMXmPnii7Q/wRP+sb19M7q/euDzaJtAoyjY09N+U6Fovxn/9cQmXlRHr1Gg5AdXV89cVXX53FccfdQe/eo/j8839z110Hcv31P1JUNABN05g9ezqTJv2eioqxVFev2ul1wuEAX345h0MPvTLp9qKiQZxxxr8oKxuJ39/A22/fwd/+tg+zZi0kJ6eMxsZNAEmtNFu/b2iI36coCpde+i4PPHA0F1+cgaKoZGQU8sc/vonHk9MVT5MhOAt8+GWQr0hBIX8h0iiza9/P/Z5Hrn6EUCBEVn4Wl9x/SaJbSNM0Zt8wm0nHTqJiaAXVG1rP/HG6nRx/yfH0H9UfRVH49v1vefCyBzn/jvMNGWaW1S6TIGNWPdka8/TTf2DDhh+5/PJPErdpvzT37bffDCZOjA/E7dNnND///B6ffvovjjnmFj744F4CgSYOO+yqdl1n/vwXCQSa2HvvM5Nur6zcm8rKvbf5fh9mzhzCxx8/zFFH/aVd59Y0jaef/gOZmYVcdtlc7HYXn3zyCPff/xuuvvorsrJK2nUeo4t6pFtJpKZIxIpDKyWobNC7FEMbNHYQ1z51Lc31zXzy4if846p/cOXsK8nMzeSDZz8g0BLgsLMO2+HjvdleDj794MT3FcMqaKhq4O0n3jZkkFlVv4pILIJVTY+3+LQaI9NTQebppy/khx9e5U9/+oCcnLLE7Vvf+EtKhiYdX1w8hNra+Hzwn39+nxUrPuMPf3Bw/vlWrrsuvrfKX/86lsceSw4rEO9WGjnyiFYtK9uzWGz07j2aqqr4k5CZGR+o29i4Oem4xsbNiUG8P//8Pt9//yrnnvsM/ftPpE+fPTn11Aew21189tnj7X4+jMzmDhOSXa5FClPDffQuwfAcLgeFvQvpN6IfZ1x/BhaLhXn/mwfAz1/9zIofVvCHff7A+RPO57pjrgPgr2f8lcdmPrbDc1YMr2DLWmOuqhqOhVldv1rvMnpMesQ1oLoa6uu79xqapvHMMxexYMGL/OlPH5Kfnzy9Ly+vguzsUjZvXpx0+5YtSxg2LP5p4OST/85RR92UuK+hYQP33HMIv/vds/TtOyHpcdXVK1my5AMuuODlXdYWi0VZv/4Hhg8/HID8/L5kZhbz88/vJaZ1+/2NrFz5BZMnnw9AKBQfN7L9jCpFUYl114jpHmbLa5a5SiKlBfyZqDYHMQnk7RaLxYiEIgCcfPnJHHX+UYn7GqobuOfCe/jdX3/X5hTtrdYtWUdWfla317q7ltUuozK3Uu8yekTaBJmlS7v/Gk8//Qe+/PIpLrjgfzidGYmxJi5XFna7C0VROPjgy3nllZmUle1B796j+Oyzx9m06WdmzHgegNzc5E9XDke8H7egoDKpdQdg3rx/kZVVwvDhrZtEX331Rvr124uCgv74/fW8/fbt1NauZt99zwXi418OPPASXn/9JgoLB5Cf35f//e86srNLGTXqaCDePeV25zB79plMm3Y9druLuXP/SXX1SkaMmNalz50uFI2gvUXvKoToFE1TcGl98Sk/612KLgK+AFVrqxLfV6+vZu3itXiyPHiyPLz+r9fZY9IeZOVn0VzfzIf/+ZD6qnrGHBSfsJBbnJt0Poc7PoO0oKyAnKL4WMDPXv0Mi9VCn8Hx1+dv3/+WeS/P47fX/rYnfsTdsrZxLcFIEIfVseuDU1xaBBlNi0+77m4fffQgAP/3f/sn3X7mmY+xzz7TATjooEuIRAI899yltLTUUla2B5dc8g4FBR1LzrFYjM8+m83ee09HVVtPAfT56njiid/R2LgJtzuHPn3GcMUVn1Ja+mu31iGHXEEo1MKTT56Hz1dP//778sc/vplYQ8brzeePf3yT//3vGu666wCi0TAlJcO44IL/0bu38fqFO8qZEyBAVO8yhOi0SLAIXOkZZFYvWs2dv78z8f1zdz0HwN5H7M1pV53GplWb+PzVz2mub8aT5aFiaAWX//NySitLd3TKNr3+6OvUbKxBtagUVxTzu7/+LhGGjCimxVhRt4IhBUP0LqXbKZqmmX6HvPXr4bXX9K5CGI2rvAa/VQb6CnOwZM8liiwjIH5V7C3myEFH6l1Gt0uLwb49OVtJpAqNkNWvdxFCdBl7LH2m24r22dS8ieaQ+T+smT7IRKOwcqXeVQijcWSFiEq3kjARLVSgdwnCgJbVmv+TvOmDzJo1EArpXYUwGjVDmuCFuQSCHlR2vCqtSE/pEGRMP9hXupVEW8IOCTLtoaJhJ4ZN0bARw6rFsGpRbMRQYzEssSiqFkWJRkHTUNBAg1/+Lz7SHgAFlK3/VdDQ4mdXVTSLhZiqElUsRBWVsKISxUJYUQmjEkIhrJn+M1fnaeDQyvArK/SuRBhIrb+WWn8tua7cXR+cokwdZEKheIuMENuyeUKEiehdhkFouInhIYKDMI5oGEskjBqNQjTSbTusKtv91wLYdvUIiyUeeqw2whY7QYsVn2alBSsRlJ0+Om2EC8EuQUYkW167nNxeEmRS0saN8TEyQmzLmuVPu0XwVDQ8ShQ3ERzRMPZoGEskhBIOQ0rslKtBNIISjWAJBbEATiCxHJnVSsxmJ2yxE7LY8CvxgBNMs5acYCATxWZBU+SFT/xqbeNaxvUap3cZ3cbUQWaDbD8i2hB1mb1bSSOTCBlaCFckgDUchLDJo1skghqJ4MCHA8jYertqiQccu4MWi4N6zUHIxK03sZiCk1ICrNW7FGEgNb4aUy+OJ0FGpBWrM0IIc43+VtHIVMJkxEI4wwEsQX+3dQmlnFgUNejHEfTjAHIBzeYg5HDgszhpVBz4TdZqo0SKwSZBRvxKQ2Nj80Yqsiv0LqVbmDbIBAJQU6N3FcJobNn+lB8dY0EjmxCeX4KLGgykSPeQMSjhII5wEAeN5ACa1UbE4cJnddCAAx+tV8pOJaFA1q4GHIk0tKFpgwSZVLNxo94VCCPSnAG9S9gtGUqEnJgfZ9AfDy6YfkHuHqNEwtgiYbL4ZcyNxUrE6aLR6qYGB9EU64qKRi3YyCZMvd6lCAPZ0GTeLgrTBhnpVhJtCVtSZ4fgTCVMTiyA09+CEk6dulNeNIK1pYlcmshVVCIuN012DzWaI2VmR1ljhYTVer3LEAZS668lEAngtDr1LqXLSZARacPmDhM2+Gq+WfwSXgJN5h+gmwq0GFZfMzm+ZnIUlajLTZPNTY3iMPbaNpEcZG08sb2NTRvpm9NX7zK6nCmDjN8PdXV6VyGMxpoRNOS062wlRHbUj9PfAhEjVigA0GJYfM1k00w2ClGXm2a7mxpchpsJFQ56JciIVjY0bZAgkyqkNUa0yWWc8TEeJUpBtBmnvxkiqT78OB1pWPwtZPlbyAJiLjcNjgy2aE5DjF6KRK1YtQwiSpPepQgDWd+0Xu8SuoUEGZE29B8fo1GghMgONGIJmH0tm/Si+n3k+H3kWKz4PZlsUT26T+u2UUQECTLiV/WBenxhH26bW+9SupQEGZEWrM4IEZ0mXtuJUaT58PgapevI7KIRXI21lFNLxO2lzualBp0WIQvLOBnR2samjVTmVupdRpcyXZDx+aChQe8qhNHYMgM9HmOylRB5oWZsvmZkunT6sfqaKaCZfJudFlcWmxRXj856CoczJMiIVjY0bZAgY3TSGiPa5OqZbiULGkX4yfA3ooT07soSRqCEQ3jDVfRXVEKeDKqtHhq17l+xLhK2YdE8RJWWbr+WSB1mXE9GgoxICxFr94YKOzFKY804WxohZuwp3kInWgx7cwOlNFDsdFHrzKJa695uJztF+JHdsMWvGoINtIRa8Ng9epfSZUwXZDZt0rsCYTSqNUa4myZe25QYpdFmXM0Nsk2AaDc14Cc/4CfX4aLGmdVt42iUaJYJX+VFZ21q3mSq7iVT/ROPxWR8jGjN5gnT1e0xVjR6ac24muplg0ax29Sgn4KgnzyniypnDnVd3OUUi3hM9iovukJdwFwLrZnqn3hDA2gyplJsR3V13W7XFjR6aS24m+ulC0l0GTXgpyjgJ9/ppsqZTX0XBZpw2AnmW5FedFJ9oF7vErqUqYJMfb3eFQhDsne+W8mCRgkteJvrISoBRnQPS8BHccBHgdvDFnsWDZ0MNNGoBRUnMYyzGKTQnwQZA5NtCURbYtbdDzIqGqWaH29LHURlBV7RMyy+Fkp8LRS6vWyyZ9Gk7f5LtU3LJajILAjxq4ZAA5qmoSjG2lpjd5kqyEiLjGhLWNm9rqUiJUBOc41sISB0Y/E108vXTMSTwVpbNsHdWC3YomWCBBmxjagWpSnURKYjU+9SuoQEGWFqFkeEKB0bjOtRovQK1qL6ZRsBYQzWlib6qj6aMnJYjxs6sLCeFs0AA2/ULfRR568zTZAx1T9vCTJiezZP+7uVVDTKtSZ6N2yQECOMJxYlo6GaQf4tZCnt/3cdjZhrXx3RNcw0TsY0LTLNsomwaIPiaN8Lfh5BCnw1EJa9kISxKaEAJaEN5HqzWWvJ3OW2B5GwA1w9VJxIGRJkDEgG+oo22Xc+PsahxCgL1WPzyS7BIrU4muvpb22mzpvHZm3Hc6xjMQWrlkFEkX/j4ldmWkvGNEFGupVEW6I7nLGk0QsfGY11sh6MSF2RCDn1m8l0udngyKVFs7R5mJVsIkiQEb+SFhkDkiAj2hJpY8/rLCVMsa8WJSRrawhzsPh99A4E8HmzWad6iW3X3aRqno6MDxZpIBQN4Qv7cNtSfwyVaQb7SteS2J5iiRHbZsaSAlTEmiip3yAhRpiPFsPdVMsA36bWg4FjMkhGtGaWVhnTBBlpkRHbszp/bY3xKFEG+jfjbKrVsSIhup8SDlFSv5HeWjMQ37NFi3XvLtsiNUmQMZBAIP4lxLZUR3zsSyk+ejdIK4xIJxqexhoGBqtxKTGiUQkyorU6vzm6MkwxRqZJxrCJNthtIfqEa7D4mvUuRQhdqAEf5aEg9dnZbNK7GGE4TSFzvHmaokUmGNS7AmE0pY4Ag2PrJcQIEYuSXVvD2JALq4z4FdsIRszx5ilBRpiKgsYIdwMjgptRO7D6qRBml+2Lsa/PRjad21FbmEcwao43TwkywjRcligTnVso9dUDEEXWhxFiq5jiwBEKMb4xQqXMYhJIi4yhSJARhfYgE9mIZ5tR322tISNEuophB0CJafRv9DM27DLHG4DYbdIiYyASZNJbP1cLo8KbsUSTW2AimgQZIbaKasldSnktfiYGHDg0U7wNiN0Q02JEYqn/OmmKf8ESZNLXSHc9A/zVKJrW6r6oJl1LQmwVi7WepOoOBJnoU2XcTBozQ/eSBBmRkqxKjL1cVZT4Gtq8X7NqaLQON0Kkq+gOVtuwhSOMa4xSqslaM+nIDN1LpggyoZ1vcCxMxq1G2Ne2mSy/b4fHaKqEGCG2pWk7nnqtxmKMaAwySAYBpx1pkTEIaZFJH/n2EPsom3DsIr1qigQZIba1syATPwAqGv2MCbuQX5/0IS0yBiFBJj1UuFrYM7yp1aDetkiQESLZLoPML/Jb/EwMOrGb4+1B7EIomvpdGqb4lypBxvyGuhsZtINBvW2RriUhkmkdWNXXEwgwscWCG0s3ViSMIBBJ/T3oUj7IxGIQlgVcTW2Eu4Hevg5ubiYrsQuRpJ2fARLs4TB7tShkaKbYkk/sgLTIGIAM9DW3Ue66xEq9HSFdS0Ika2/X0rZs4QjjWyBLk+nZZiWDfQ1AupXMa4yrliJf4+49OOX/ZQvRtTraIrOVNRJhXHOMXFlrxpRksK8BSLeS+ShojHPVkO/f/S3mpUVGiGS70yKzlSUaZc+mGPm/bHMgzENW9jUANeV/ArEtFY0Jrhpy/c2dOo8EGSGS7W6LzFaWaJTRjRGKkYXzzERVUv9NNOV/AosMqjcNi6Kxl7OKLH9Lp88lQUaIZLEuGAGvxmKMbAzJKsAmIkHGACTImINVibG3YwsZAX+XnE+2JxAimRbrmt8JJaYxvDFEb83ZJecT+rIoqf8mmvJBRrqWUp+KxnhHNZ5A161nIC0yQiTryDoyu6JoGkMag5RIy0zKkxYZA5Agk/rGumq6rCUmQdaRESJJZ8fIbE/RNIY3hcnTZABwKrOo0iKjO+laSm2j3XXkdMGYmO0pnZihIYQZqd3wK6HGYoxujpK5g521hfFJi4wBSJBJXcPcDRTu7joxu6DEJMgIsa3uer+yRKOMa0a2M0hRMkbGAKRrKTUNcDVTthsr9raXqsk/DCG2ZenGcWPWSITxPotsNJmCpEXGICTMpJZyp49+/ppuvYa0yAiRTOnmjVQdoRB7+W1YZYBaSpExMgYhQSZ1lDoCDApWd/t1JMgIkczSAzvCu4JBJgQd5nhjSRPSImMQMk4mNeTZQwwPV6F09fSJNqhRU/zTFqLLKD20JIHXH2Bs2NUj1xKdJ2NkDEKCjPG5LFFGxapQYrGeuWAPXUaIVKEqPfdLkdPiZ1hUwkwqkBYZg5CuJWNT0RhrrcYa6bnNyaRFRohkSg8GGYCyJj+9ZPVfw5MxMgYhQcbYRrvqcAe7btXedon27OWEMDpVh9WuhzaFZI0Zg5MWGYOw2fSuQOzIAFcz+f6mHr+ugmKKX1AhuoqqQ7pXYzHGtCjYZIFKw7KqqR80TfFK7/HoXYFoS6E9SN9ArW7XV83xz1uILtHTXUtb2cNhxoZkTyaj8tq9epfQaaZ4pfem/t+D6bgtUUZGe2aG0o4oinwKFGIrVdGvvzXTH2BE1K3b9cWOSZAxCAkyxhIf3FuFJarvQBWLLJkuRIKi88Cx0iYffWTwr6FYVStOa+r/nUiQEV1ujKsWVzCodxlYZZChEAkW9P+dHNQUIhsZ1GgUHps5xmVIkBFdqtLVTK6/We8yALBqEmSE2Mqi+fUuIb5bdgtYZPCvIZihWwkkyIgu5LVG6Bes07uMBGtUgowQABYVFHpuHaedsYfDjIykfneGGUiQMRC3G2Rcp74UNPa0VKP21Mq97WANS5ARAsBmNc7vJUBhiyyWZwQSZAxEVeNhRuhnqLvREONitmWJyGBfIcB4QQZgSHMYh2aKt6CU5bHLGBlDke4l/eTbQ5T5GvQuoxVLUIKMEAA2izG6lbZliUYZHZb1ZfQkLTIGI0FGH1YlxgitGtBvvZgdUWIKVkW6l4RQVWO1lm6V5fPTX5PNJfUiQcZgJMjoYw9nPfZwWO8ydsimyFRPIayE9C5hh/o2BcmQGYa6kCBjMLJNQc8rc/p12UepI2yyZoUQWJUe3rS1A9RYjNEBCzrsaZnWnFanKfZZAhMFGWmR6VlONcrgcI3eZeySNWaOX1QhOsOi+fQuYadcwSDDYjJjoyeZpTUGJMiI3TTSUaf7FgTtYYnKgF+R3mxWMOIYtu31avaRpUkLak+RIGNAmZl6V5A+ShwBcvwtepfRLrKWjEh3Rpx63SYNRshMwx6TYc/Qu4QuY5ogY7dLq0xPUNEYHKvVu4x2s4YkyIj0ZrMab+r1jngCAfrGZBZTT8hz5+ldQpcxTZAByM/XuwLzG+xqMvQspe2pIRWLIp/yRPqyW4w59XpHKltC2M311mRI+W7zvGGa6l+LBJnu5bFEKAvU611GhzkVWQpdpC8rxp5ZuD1LNMowWSivW1kUC9nObL3L6DISZES7jbTVoWjGHzS4PUdUXhRFelIAe4oFGYjvxZSn2fUuw7RyXbmoinne/s3zkyBBpjuVOf1kBow9hXNH7GF5QRTpyeGIkQozltoyLCA7AXcXM3UrgcmCjNstm0d2B4uiMTCSOgN8t2f3SZAR6clpT53xbNtzBYMMkIG/3cJsQcZ0Uzry82HNGr2rMJehzgZs/tSZ+bA9JaJgV+2EYsZdpt1IfH4fjz79KHO/mEtdYx0D+g7gorMvYkj/IQDccu8tvPnhm0mPGT9qPLdfd3urc4XCIc6/8nyWrVrGI3c8woC+AwB47NnHmP2f2a2OdzqcvPXUWwC88s4rvPXRW6xcsxKAQf0G8bvTfseQAUO68sc1Nbuamq2oW5W3hFiboRIgRaaQpwgJMgZXUCBBpit5rRFKAo16l9FpDs1ByMD7zRjJbQ/cxso1K7nmj9eQl5vHOx+/w/+74f/x+N2PU5BXAMD40eO58g9XJh5jt7Xd6vXQvx8iLyePZauWJd1+0pEnceTUI5Nu+9OsPzG4/+DE9wsWLuDAfQ9k+KDh2G12nnrpKS678TJm3z07UYfYObvSnKo9S8DWgb8uvrH59S7FNCyKxVRTr8FkXUsARUV6V2AuQ2wNKTnAd3uOmAz4bY9gMMjHn3/M78/4PXsM24OykjLOOuksehX34n9v/S9xnN1qJy8nL/GV4W29uNbn337OV999xQVnXtDqPrfLnfT4uoY6Vq1bxeEHHp445rpLruOYQ49hQN8BlJeVc8X5VxDTYnzzwzfd88ObjEUFi5YaC1fuTL7PL5tKdqECT4GpBvqCCVtkiopAUcAE7726y7KFyU2RFXx3xRa0gczC3qVoLEo0Fm3VwuKwO/jh5x8S3y9YuICjzjqKDG8Go4eP5txTzyUrIytxf219LXc8eAc3/fkmHI5dh8hX332V3qW92WPoHjs8JhgKEolGyPTKMt7t4XAYfwuRdtFgSMTGl7bU7d42kiKP+T7tmyuWATYb5OToXYU5DLY0kNLt0tuw+WQPl/Zwu9wMGzSMfz//b6prq4lGo7z90dssXLKQmrr4JqHjR4/n6j9ezZ2z7mTG6TP4btF3XHHTFUR/2XtL0zRuue8WjjzkyKSuoh0JhoK8O/fdpNaYtjz0xEPk5+QzZuSYzv+gacBhTa2F8HYmp8VPtuxk3yWKvOYLMqZrkQEoLoba1J1kYwi5thDZAXO0xgAoMQWH6iAYM8+Le3e55o/X8Lf7/8ZxvzsOi2phQL8BHLjvgSxevhiAA/c9MHFsZXklleWVnPKHU1iwcAFjRo7hv6//F7/fz2nHnNau6839Yi4+v49D9z90h8fMeWEO7897n3tuuAeHXboJ28Nh8ZvlcwgAg0NWPk/hWVhGYcYWGVMGmaIiWLRI7ypS22BLPZjsNcOJkyASZHalV3Ev/v6Xv+MP+PH5feTl5DHr/2ZRWlTa5vGlxaVkZWaxftN6xowcw/wf5rNwyUIOPvngpONmXDGDgyYdxNUXXZ10+2vvvcbeY/YmNzu3zfM/879neOrFp/i/mf9HZUVl1/yQacBG6g/S31aWz0++3U61DNrfbZmOTFw2801pN22QEbuv0B4kI2C+WQKOsANk26V2czlduJwumpqb+GrBV8z47Yw2j9tSs4XGpkbycuIzIf54zh8559RzEvfX1NZw2V8uY+afZjJkYPLU6Y2bNzL/x/n89cq/tnnup156iif/+yS3X3d7u7qpRJzdBqpmvtA+KKBSLWPddluxt1jvErqFKYNMZiZ4PNBinp6RHjVIqde7hG7haHZA1q6PS3dfzv8SDY0+pX1Yt2kdD/37Ifr06sPhBxyOz+/j8f88zqS9J5GbncuGTRt46ImH6FXci3GjxgFQVJD8ScLljH8CLC0upTCvMOm+199/nbycPCaMntCqjqdefIp/PfMvrrvkOooLihNjdFxOF26XrHy5M26n+UIMgDcQoNjpYJO0rO4WCTIpprxcupd2R6nDjzsY0LuMbqGGVZyqk0DMnD9fV2n2NfPPOf+kqqaKDG8Gk/eazLmnnovVaiUajbJ89XLe/PBNmn3N5OfkM3aPsZxzyjk7XEtmR2KxGG988AaH7n8oFkvrprL/vfU/wpEw199xfdLt00+czlknndWpn9HsXJZGU42P2dZAP2wyX+9It1NQKM8q17uMbqFomjknKq9bB6+/rncVqWeyfRPOkHk/7TTnN1NDjd5lCNFtVAXKMpegmHg13IVZTtYp8oGkI4o8RRw1+Ci9y+gWppt+vVVpKdhli50OKXX4TR1iABw+mfEizM3tipo6xAD0DZj75+sO5dnmbI0BEwcZVYWyMr2rSC19lSa9S+h2Np8Nq2raHlUhcNnMPzjQHQxRiHxS7YiK7Aq9S+g2pg0yABUVeleQOnJtIbwmnKnUFg8evUsQots4SI9FtPqFZApie2U5ssh2ZutdRrcxdZDp0yfeMiN2rb/VXGtO7IwrKCMFhTk57RoW0mPsSJbPT6Z556t0KTO3xoDJg4zdDiUleldhfG5LlBy/T+8yeoy9yY6ConcZQnQ5tzM9QsxWA8KybUF7SJBJcdK9tGv97U2Ydq5mG5SYgluVdUiE+ThNugbUjuT5g9jN/zbWKS6ri0JP4a4PTGGm/xdQbt6B2l3ComgUhpr1LqPHuSLSvSTMxWIBKw16l9GjlFiMyqgs9bsz5dnlKIq5W6BNH2S8XsjP17sK4+rrbMHyy67F6cTZIi9+wlw8zlBadpiWBEIo6dOg3GFmXQRvW6YPMiCtMjtTFjX/lOu2WIIWHKqsKSPMw2VNv5ZVAFs4Qhnyu9wWq2qlV2YvvcvodmkRZGScTNuKHEEcofTdSdYb8+pdghBdwqKCI41XrC4PpmNb1K6VZZalxbpZaRFk8vLiXUwiWR8lPT/BbeVulAG/whw87iAK6ddFvJUnEMArU7FbMftspa3SIsiAtMpsz6rEyA6lz5TrtqghFY8qi+OJ1Oe1pMcieDtTHpWVfreloNAnq4/eZfSItAkyAwfqXYGx9Hb6UWOyX4knKEFGpDaHXcOWZrOV2lIYCOtdgqH0yeqD05oekxrSJsjk50NBgd5VGEepZv79WNrD2ejEoshS5yJ1ZTjldxnAHg5ToEmrzFZDC4bqXUKPSZsgAzBsmN4VGIPLEsUbSK8VQHdE0RS8yAAqkZoUwKVs0bsMw+gdlXEyAJmOTHpn9da7jB6TVkGmshIcMkuPCkcL6bSS7654WqR7SaQmjzuCSvrOPNxerj+YXm9qOzAkf4jeJfSotPo7t1hg0CC9q9BfUUSaordl89lwqunRlyzMxWOr17sEQ7FEo/TS0vvTqkWxMCg/vd7o0irIAAxNn27DNuXYQmm9dsyOeCPSvSRSi9UCzjReO2ZHeoXT7m0tSb+cfmkzyHertPsbz8yEsjK9q9BPuU1aY9rianShKmn36yBSWIbbj3QRt5bpD6T1RpLDCtNvMGha/m2na6uMgkZeSIJMW9SIikeRsTIidXgs1XqXYEiKplEeS8/upXx3vul3um5LWgaZ8vL0XOm30BHCGknf1T93xeOXICNSg8sRwyJLKOxQUSg9W6rSacr1ttIyyCgKDB6sdxU9r9ji17sEQ7M32WUjSZESMp31epdgaO5gEGua7QVut9jpn9tf7zJ0kZZBBmDIEFDT7KfPiUiQ2RkFhaxwlt5lCLFTNhs4kLVjdkbRNErTbPbSwLyBabFBZFvS7K38Vy5Xeu2/5LFEZLZSO7jqXdhUm95lCLFD2e6GNGtr2D2FkfR6ltK1WwnSOMhAeq30W2qXlXzbRYPsSLbeVQjRJqsV3GzSu4yUkJVGey+VZpSS7czWuwzdpHWQKSmBnBy9q+gZ+Zp0K7WXq96FVUnPJlphbNnuJmTKdftYoxFySY/W1XRujYE0DzIAo0frXUH3U9HICEmLTHspMYWsmIyVEcZisYBbkdaYjiiJmj/IZDuz6ZvdV+8ydJX2QaayEnJz9a6ie5U4AyixmN5lpBRPvUd2xRaGku3xoSDLJ3REXsj8z9e40nEoSnqNB9pe2gcZRYFx4/SuonsVqdKt1FFKVCFLk1YZYQyqCm51o95lpBxXMIhDM+/bXIG7gL456d0aAxJkgPgCeYUmXgwxOyRBZnd4672ybYEwhGxPAFVLn8GrXakX5p2GPa6XyT+Ft5O8Sv9i/Hi9K+gemdYwtkhE7zJSkhJRyCRT7zJEmlMV8FikNWZ35Zv05a/EW0JZZhpvHLgNCTK/KC2FXr30rqLrFdiCepeQ0rwNXhRZtUPoKNMTRNXk93h3eUPmbMka38ukn753gwSZbZixVSYHWQSvMywhC5mqtMoIfSiA1yozlTrDFo7gNNlbXZ+sPhR5i/QuwzDM9bfbSQUF5lvt1xuVT3KdlVmfKWNlhC6yMoJYZA2oTivU7HqX0KXGlcrYmG3Jq/N2xo2Lz2QyA6sSk20JuoAaUmUGk+hxFhUy1fV6l2EKuVHzvNVV5lSS587TuwxDMc/fbhfJyYH+JtlAtNAuIaarZNRmyGq/okdlZzSjSNdwl8iImGM9GVVRGVs6Vu8yDEeCTBvGjjXHztg5FulW6ipKTCEnmib7WQjd2W3gYYPeZZiGKxRCMcHODgPzBpLllNbh7Zng7brrZWTA4MF6V9F5WTEJMl3JVevCoZp3TQphHDmeOhRkNe6uosQ08pTUHidjUSyMKRmjdxmGJEFmB/bcM77TbCpzh6VZuispKOT4pVVGdC+XM4ZT26x3GaaTF0vtLUeGFgzFY/foXYYhSZDZAbcb9thD7yp2X6Y1jCVqjn5hI3E0OfCqXr3LECaW45Tp1t0hO4UXxnNanYwuSYMdjneTBJmdGD06Pvg3FclCeN0nuyFbpmOLbpHlCWLTGvUuw5Q84dRdGG+f3vvgtDr1LsOw5NV4J1QVJk1KzenYmUrq/tIanSVoIVvL1rsMYTIWC2Ta1uldhmnZwhFsWuq9mPfJ6kP/XJNMpe0mEmR2oagIhg3Tu4qOc8ckyHQnb40Xu5ragweFseR6G2VjyG6Wpdj0LqFDbKqNffvsq3cZhidBph3Gjwdvig2LcETkBbE7KZpCbjBX7zKESTjtGi6Zbt3tMrXUGvA7vtd4vPYUe/PRgQSZdrBa411MqcKiaLLjdQ9wNMjAX9F5CpDr3ihbk/YAbyx1nuVibzHDClOwO0AHEmTaqawMBg7Uu4r2ybRKa0xPyanNwaKk1qc8YSw5mS0ywLeHuKOpsSqeRbEwqTyFPj3rTIJMB+y9N7hcelexa1kWaY3pKWpEpSBUoHcZIkU5HRoZigzw7SnOFNmqYM+SPcl2ZutdRsqQINMBDgdMnKh3FbvmVaVFpic5Ghxkqpl6lyFSjKpAvmsdkBqtBGZgD4cNv1VBniuPPYpTeBEzHaT42rU9r18/KC+H1av1rmTHPAac+bCpbhO3vnArHy78EH/IT0VBBbefeTsjK0YC0BJo4W8v/o23F7xNXUsdvfN7M33KdE6ffHqrc2maxvR7p/PRwo94+PyHOWTUIQA89+lzXP745W1e/+vbvyY/Mx+AzxZ/xk3P3cTSjUspySnhwsMv5IR9Tkgc2xxo5v/+93+8veBtqpuqGdZ7GDNPmskeFTt+ccmuycaf5ycss8VEO+VmNmHRWjp1jk3Vddz6zxf48MuF+IMhKkoLuP3yMxk5qKLVsVffPYenXv2Y684/gXOOO6jV/cFQmKMvupWflq/jtYeuZVj/3on7PvpqIXc9/gpLV2/AYbcxfsQArvn98fQuzu9U/T1N0TS8WGjCmC0zCgqTKybLOlUdJEFmN+y7L2zcCCGD7gDgjBrrzbShpYHjbj+OvQfuzeyLZpOXkcfKLSvJ8vy6+dlNz93Ep4s/5a6z76Isr4y5i+Zy3dPXUZRdxMF7HJx0vkffexSljcV9fjP2N0weNjnptstmX0YwEkyEmLXVazn7vrM5bdJp3HPOPcz7eR5XPnElhVmFicf++d9/ZsmGJdx51p0UZRfx4hcvcvpdp/POrHcozilu82dUogr5vnw2Ojd26rkS6cHtiuJhfafO0dDUwnEX387eowYy+5aLyMvKYOX6LWRltF7G/s1P5jP/pxUU5WXv8Hy3/PMFivKy+Wl5clfX2o3V/O76Bzj3+IO45+pzaGrxc+OD/+H3sx7itYeu7dTPoIcszUqTYswgM6JoBPnu1AqHRiCxbzd4PDBhgt5V7JjDYDOWHnzrQUpzSrlj+h2M6juK3vm9mTR0EuUF5YljvlnxDcftfRx7D9qb3vm9OXXSqQwpG8J3K79LOtfCtQt55J1HuO2M21pdx2l3UphVmPiyqBY+W/wZJ008KXHMkx89Se/83lx7wrX0L+nPmVPO5LA9D+PRdx8FIBAK8Ob8N7nquKuYMHACFYUVXPqbSykvLOfJj57c6c9pb7aTrWR34pkS6cCiQp6j8+NiHnzmLUoLcrjj8umMGtyX3iX5TBo7lPLS5DFbm6rrmHXfM9xz1TlYrW0PTP/gyx+Z+80irplxXKv7fli6mlgsxmVnHUV5aQHDB/ThvBMOZtHydYRTZMzJtjI0Y77tZToyGVs6Vu8yUpIx/0ZTwJAhUFqqdxWtuS1RlJixds199/t3GVE+ggsevoAxl43h8JsO5+m5TycdM6bfGN797l021W1C0zQ+XfwpKzevZL+h+yWO8Yf8XPzoxdx4yo0UZhXu8rovfP4CTruTw/c8PHHb/BXzmTg4eaDTpKGTmL9iPgCRWIRoLIrDmrzLtdPm5KvlX+3ympnVmbJDttipvIx6VM3f6fO8+9n3jBhYzgU3PsyY4y/j8Bk38fRrc5OOicViXHrrY5x34lQGVrT9glVV18hVdz7BXX8+C6ej9SKPIwaUo6oqz731KdFojMZmPy++8wX77jkY2w6CkZF5jPXymDCpfBJWVTpJdocEmU6YMsV4s5jcFuN9QlpTtYYnP3qSisIKHv/j45w+6XRmPTuL5z97PnHMrJNn0b+kP3tduRcDLhjA9L9P58ZTbmTCwF+bvm78z42M6TeGqaOmtuu6z857lqPGH4XT/useJVWNVYlupq0KMgtoCjQRCAXwOr3s2W9P/v7639lcv5loLMqLn7/Ityu+paqhapfXVDSFvOY8FFkVRLQhwx3BRddsCrlmYxVPvvIRFb0KefyWP3L6byYx6/5nef7tzxLHPPjMW1gtKmcdc0Cb59A0jctum81pR0xqc1wNQO+SfP5968Xc/uhLDDzsD4w8+hI2Vtdx33XndcnP0dPsMeON9h1dPJrSDAN+Mk4REv86weOBAw+E114DzSC/G07VeEFG0zRGlI/gimOuAGB4n+Es2bCEOR/N4fi9jwfg8Q8eZ8HKBTxywSP0yuvFl0u/5Pqnr6cou4h9h+zLO9+9w2eLP+O1a15r1zW/Wf4NyzYu466z7upwvXedfReXP345E/48AYtqYXif4Rw57kh+WPNDux5v89nIdedSQ02Hry3My2qBHNuaLjufpmmMGFjOFeccA8DwAX1YsmoDc175iOOn7s0PS1bz2Ivv89qD17Q5pgxg9ksf0OILcMEph+3wOltqG7jqzic4bure/GbKOFr8Ae6c/QoX3PAwT952yQ7PbVQ2g7VYl2WWSZdSJ0mQ6aTS0vgWBl98oXclcU4DDmIrzCpkQMmApNsqSyp5Y/4bQHxcyu0v3c7D5z/MASPinxyHlA1h0dpF/OPtf7DvkH359OdPWV21mpGXjkw6z/kPnc+4AeN49v89m3T7s/OeZWjvoYwoH5F0e0FmAdWN1Um3VTVWkeHMSLTclBeU85/L/oMv6KM50ExhViF/+Mcf6JPfp90/s7fGi6/Ahz/W+S4EYQ753hoUum6GQGFuFgPKS5Juq+xTwhtz492kX/6wlJr6JvY59arE/dFYjJsffp5/vfA+8+b8lU/n/8y3P61g4GF/SDrPkRf8laMOHM+dfz6LJ/73IRkeF1ed9+v4mbuvOpu9T7mS+T+tZM+h/brsZ+oJ1qhxgozX7uWAvgekXBg0GgkyXWCPPaCqClas0LsScCjG+SXdakzlGFZsTn5yVm5eSa/cXgCEo2HC0XCrX2ZVVdF+aeo6/9DzOXnfk5PuP+TGQ7juxOs4aGTyVNKWQAuvff1aogVoW6P7jebDHz9Muu2Tnz5hdL/RrY51O9y4HW4aWhr4eNHHXHXsVa2O2SEN8hry2Ji5kahmvHApelZORgAHu+6a7IgxwypZsXZz0m0r122mV1F8D7BjD9qLffccknT/GVf+nWMOmsAJh+4DwKw/nMxlZx2VuH9zTQNnXHkP9137O0YN6QuAPxhCUZN/Ny1qfFSCZpSm6A6wxIzx+2hRLEytnIrT6tz1wWKnJMh0kcmTobYW6uv1rcNuwPURzjnoHI7723Hc//r9TBs7je9WfcfTc5/mltNvASDDlcGEgRO45b+34LQ5Kcsr4/Mln/PC5y9w7Qnx6Z1bZyJtrzS3lN75vZNue/XrV4nEIhw94ehWx58++XT+/eG/ueW/t3DCxBP47OfPeO2b1/jXhf9KHPPRwo/QNI3K4kpWbVnFX//7VyqLKzlh4gmtzrczlqCFQn8hm5yb0GTRs7TlcUXJUFd1+XnPOe4gjrv4b9z/1OtMmzyW735exdOvz+WWS+NrL+VkecnJSt4LzGq1UJCbSWXv+DICW0PPVm5XfKB6n9ICSgpyADhgwgge/e973PPEqxw5ZRwt/iC3PfoivYryktaaSRVKTMOmKYR1Xhlv3z77ylTrLiJBpovYbDB1Krz4IoR1XMbFbsBP/3tU7MHD5z/MbS/exj2v3UPv/N5cf+L1SUHj3nPv5bYXb+OSf11CfUs9vXJ7cflRl3P6pNYL4u3Ks/Oe5dDRh5Llzmp1X+/83vzrwn/xl+f+wmPvP0ZxdjG3/vbWpPVnmvxN3PbibWyq30SWO4vD9jyMy46+DJvF1uFa7M128ux5VKvVuz5YmI7NBrmOVd2ymuwegyt4+Ibzue2RF7nnidfoXZLP9eefyNEHdu3aEPuMHsw9V5/Dw8++xcPPvo3LaWf0kH48fssf25zllApcioUw+i1TMTh/MIPyB+l2fbNRtFRsGzSwlSvhnXf0u/5Ex2a8wYB+BYg21RfW0xBr0LsM0YNUFUoy12HVmvUuRWxnQZaDzUpQl2sXuAs4ctCRWNTUm7puVDL9uov17RsfM6MXmwFbZARkVWXhUg02V190q8LMGgkxBuXUaXkEp9XJwZUHS4jpYhJkusH48fotlmekEfniV4qmkF+XLwtepYm8TB8OrWsH94qu49B6PsgoKBzY90C8du+uDxYdIkGmGygKHHRQfJ2ZHr0uGpaotMgYlRpWKWwplA3hTC7DHcGrdN16MaLr6RFkxvUaR6/MXj1+3XQgr6jdxOmEgw+O95P3FLsqrTFGZ2uxkR+RmQpm5bBr5NhW6l2G2AVbD48MrciuYFTxqJ69aBqRINONCgth4sRdH9dVLLKmUkpw1blkc0kTsligwL0GxYBLIIhkSg/Occl2ZrN/xf49dr10JEGmmw0ZAnvu2TPXUnVeF0G0X2ZVJm6LW+8yRBdRgMKMLVi6YDNI0f3UHnqp9Nq9HD7gcOyW1JymniokyPSAsWNh+PDuv44qi66lDAWFvJo82SnbJAqyG7BrtXqXIdqpJ1pkXFYX0wZMk8G9PUCCTA/ZZx8Y1M3rH1mkRSalqBGVwrpCCTMpLj+rCZe2Ue8yRAeo3by3kcPiYNrAaWQ5Wy/KKbqeBJkeNGkS9OvG/dXkLzP1qGGVgvoC7Ko0PaeivMwWPKzXuwzRQd3ZImNTbRw24DByXbm7Plh0CXnv60GKAgccAL27aXsSaZFJTZaQhcKGQlljJsXkZPjxKmv1LkPshu4KMls3giz0tN4XTnQfCTI9TFXj07JLSrrh3BJkUpYlaKG4sRiLIit+poLsjCCZ6mq9yxC7Se2GlX1VReWgfgfJWjE6kCCjA6sVDjkE8rt4OREZ7JvaLAELRS1FEmYMLssbIkuVtWJSmdoNLTL7V+xPeXZ5l59X7JoEGZ3Y7XD44ZCT03XnlL/M1Gfz2SjyFcnqvwaV4Y6QZVmhdxmik5Qu/tC3b5996Z/bv0vPKdpPXi115HTCtGmQmdk155MxMuZga7FR5JcwYzRed5Rc23KdthsUXaoLW2Qm9JrA0IKhXXY+0XHySqkztzseZnp6XyZhbPZmO4XBQhR52zQEtyseYpDuW1Poqt+r0cWj2aN4jy45l9h9EmQMICMjHmaczs6dJyJveqbiaHRQGJYwozeXM0a+fQUKspeZWcS6YB2Z4YXDGddrXBdUIzpLgoxBZGfD0Ud3rpsposlfp9k4G5wUBaWbSS9ed4QCxzLZP8lkOhtkxpSMYZ/e+3RRNaKz5NXRQDIz42GmcDeXIIjosDW96H6OJgfFLcVYFVlnpidle4Pk2pZJS4wJRXfzpVJVVCaXT2ZM6ZiuLUh0igQZg3E64YgjoHw3ZvGFYxJkzMrms1HcWIxNteldSlrIy/SRZVkpnXomtTvtazbVxiGVhzAov5v3mhEdJkHGgKxWmDoVhnZwILx0LZmbJWihpLYEp9rJwVRipwqyGvAqa/QuQ3SjmNqxiOq2ufnNoN/QO6ublmUXnSLvfAalKLDvvjCuA2PJpGvJ/JSIQmF1IR5Vprl1NVWB4uxq3MgGkGbXkTEy2c5sjhp0FPnuLl7BVHQZCTIGN3o0TJkS39pgVyLStZQWlJhCflU+mWoXLUAksFigOHMjDq1a71JED4i0cxp9sbeYowYdRYYjo5srEp0howdTwIAB8fVm3nkHQqEdHxdDARRkrYs0oEHOlhws+RbqqNO7mpRms0KRdy0WrUXvUkQPibajRaZvdl8O6HsAFlW2DDE6aZFJEb16wZFH7nrhvI72/YrUllmdSX5Mmrx3l9OuUexdKSEmzexq1tLwwuEc1O8gCTEpQoJMCsnNhaOO2vn+TF2x0JNILZ5aD8Vh2Tm7o7zuKIWu5ahaUO9SRA+L7qTVeq+yvdin9z4o8lqaMiTIpBivNx5mSkvbvj9qkb/SdORocFDSKDOa2kMB8rKaybMtRSGidzlCB+E29qWzKBYO6ncQI4tG6lCR6Ax510tBW3fO3qONLT4i0hSatixBC4VVhWQpWXqXYlhWK5Rkb8bLOr1LETryb9cik+nI5OjBR9Mvp59OFYnOkMG+KUpVYcIEKCmBDz6A4C+t4yHFgkzMTV+KppBdlY0jx0G1tZqYJqvSbuV2xcizr0LVdjJiXqQF3zZL4vXL6cek8knYLXYdKxKdoWhaF+5nLnTR3AzvvgtbtsAodx1Fvka9SxIGEHFFqMqoIhSTN+6cDB8Z6hpZqVcA8G6WgqYq7F22N8MKh+ldjugkCTImEYvBF19AZHkjfXwyHVfEaRaNurw6mmJNepeiC4sKBZnVsj6MSIipKp/nezio30EUeAr0Lkd0AQkyJlP7Yw0tnzcSi8nwJ/GrltwWatQatDRaY8jp0ChwrUHV/HqXIgzE6cglZ+8jpSvJROTdzmRyC6OUKJ/idPj0LkUYiKfWQ4m/BKuaHsPisr1BipxLJMSIBAWFvGYXRc1eCTEmI0HGbDIysEZbKAx+Qo57nYwJEAm2FhultaWm3trAaoXi7BqyLCuRFa7FVnbslGy04t3sj69hIUwlPT6epROXC6xWlEiETN9CnPbNVGsjCIflE4iIbzqZsyUHd6abamc1kZh51lHJ8oTIsq5B0czzM4nOUYCMkJvsdT4SS8dkyL5JZiMtMma0zScOe6iakvBHZLs3oLSxCJRIT45GB6XVpWQr2XqX0ml2G5RkbyHbukIWuBMJduwUb7aTs3abEAMSZExIWmTMKCMD6usT3yrEyPL9gNu2lhp1JMGgS7/ahGEoMYWsqizcXjfV7uqUnKadnREgU12LokV3fbBICwoK2T4XGRt9bXetS9eS6UiQMaMd/KLawvUU8zHNrgHUBStkZpMAwNZso7ilmKa8JuqpT4mZTQ67Rp5rEzYa9C5FGIgr5iR3YwRrYCeTHaRFxnQkyJjRLn5Rvf6luCzrqXWNxOeX5exFfEXgzOpMXG4Xtd5aArGA3iW1SQFyMlvIUNYhg3nFVhYs5DTa8FTt4t+t1RofRyhMRYKMGe1se+xfWKI+Cvyf43P2oS4ygEhE/ikIsPlsFPmLaM5tpk6tM9QWBy5HjFzXBqxas96lCAPxRlxkbwhiCbcjfGdnd3s9oufJu5cZFbR/tUp3YA1OZSMN7uE0+QvQNJmwnfY08NZ4cTld1GfW0xzTNzhYrZDjacTFBmS8utjKjp2cGgVnfQfWCurAa6NIHRJkzMjtjn/52rconqqFyfHNJ8Pqpd46lJbArlt0hPlZAhbyAnlkejKp89bhj/bs4nKqCjleHx5lPQoymFfEWbGS1WjFUxXo+DpZ+fndUZLQmQQZs8rPhzVrOvQQa6SZ/MiXZNrzqVMHEwjIPtoivpBeYUshwYwgda46grFgt15PAbIygmSq61FIvZlUonuoqGT5nWRs8qPs7vpH0iJjShJkzKqgoMNBZit7qJoiPsHv6EWdNoBwyNHFxYlU5GhyUNRUhD/bT729nnAs3OXXyPSEybRuxIJssSHi4ovaucjcFMQS7sS/C4sFcnO7rC5hHBJkzKoLPnm4gutxsoEWVyX14XKiMiA47SkouOvduBQXLbkt1Kv1RLtgDRePK0qWYzM2rbELqhRm4Ym6yN4cxurvgm7N3Nx4f6UwHXlnMqsuakJV0PD6l+FWVtHkHkRTsIRo1NIl5xapS9EUvDVe3FY3zdnNNCgNuzXDyWnXyHZV4aBWZlOLBGfMSU51DHtTF47Lkm4l05IgY1YuF3g80NLSJadTtQhZvoVkKItpcVXSGC6TKdsCNaKSWZ2J1+6lMbuRJq2pXYHGYdfIdtfj1Db3QJUiVXiiLjKro9ibu2EdIwkypiXvRGZWUNBlQWYrVYuQ4V+MlyX4XBU0RssJyRiatKeGVLK3ZJNpy6Qlq4UGpaHNLie3K0aWowa7ViMtMAKId1d6w04yq7qoC2lHZMaSaUmQMbP8fFi1qltOraDh8a/Ew0r8zl40av0IBN3dci2ROtSwSkZ1Bl7Viy/bR4O1gXAsTIY7TIa9Kj4GRgKMID4LKSPoJGNzAEu4m6f2W63tWihUpCYJMmbWQ02prsB6XKwnaC+g0VIp2x4IlJhCRr2LYkczFs9a/FYfQU2mUov4OjAZPjvezX7UWA/NTpOBvqYmQcbMerhP2BGqooAqwtZMmm19aQkVyMDgNGSzhciwbsATWI7qj4AfsqohlGGnMdeCzxpIiY0pRddyxZx4mhTcNX4UbTfXgdldMj7G1CTImJnTGd8Ju7lnl5i3RRrJiXxHNgoBZxnNlOEPZMpbl4kpiobb2YAnthZXcAO0scSMvSlEfhNEbSrNBS5aXGHCbR0oTMOKFW/AjqcmhDWg40akEmRMTYKM2RUU9HiQ2UpBwxVYi4u1RC1uWhx9aQ4XEQ7bdKlHdD2nw4dH2Yg7uBrV375QYgnHyNrgI4t4K01ztgWfPURUtiEwBQUFd9SJpyGGqy4I9HDrS1skyJiaBBmzKyiAlSv1rgJL1EembyGZLCTgKKZZ7YMvkC2bVKYgmy2Ex7YFT2g11mDnQrK9KURuE+QoEMhx0pKp4LNI11MqsmHH67Piqe6BwbsdYbXKrtcmJ0HG7EpL9a6gFWdwE042EVNs+F298FOEP5hJLCaD8YzKYonidtTiCa/BEa5us+uoMxQNXLUBXLUQs6r4ch20eDUCSvfu6yQ6x44dd8CKqyGMvTkERtwbq7QUFPnAZGYSZMyusLBDO2H3JFUL4/GvwsMqNFQCzhJ8SjH+UI4MEjYAmy2Ey1aPK7oZR3ATiq/jK/fuDjUSw7slgHcLRJxW/Nk2/C6NgBqUlhoDcMbsuAJW3HUhrAGDhpdt9emjdwWim0mQSQd9+sDPP+tdxU4pxBLTuAGC9kJ8lhL8kTwZU9NDFEXD4fDhUqtxhTZgCzd2ectLR1kDETI2RcgANFUhkO3E71Xw28JEjDD2Ig0oKDijdtw+FVddEEs4BcLLtsrL9a5AdDMJMukgBYLM9hyhLTjYQg4QtmXjtxYT1HIIhj3SWtOFLNYILlsjLm0LzuB61IBxw4ES0xLdTwBhty3eWuOMElSC0lbThezYcYasOHwxnHVB1FiKdvHl58e3ahGmJkEmHZSVxbewj6bmrBBbuB5buD7xfciWQ9BaSJAcAiGvBJsOsNlC2K0tOKjHEanGHq41xKSS3WHzhbH5wmQSH1cTzLQRdFsI2mOElBAxeqYrLNUpgF2z4whbcbZEcTSGUSMp1uqyI9KtlBYkyKQDqzU+4G3tWr0r6RL2cB32cB0Zv3wftmYStBYSUHIJhr1EItIVBaCqGna7D4faiCNWhz20BUs4qHt3UXdQIzFctcFEa40GhL02gl4rQQeErBFZs+YXKip2zYYjZMHREsXREEKNmSS4bE+6ldKCBJl0UV5umiCzPVukEVukEe8v38cUG2FbDmFLFiElg3DUQzjiMHXLjcUSxWYNYlP92GjEEanGFq5DCaRnh4sC2JvD2JvDicAbtamEMuwE3Sphq0bYEiVC2LQDiBXAig171IotrGALxLC1hLEFokCKdhV1hNst68ekCQky6SKNmlhVLZwYY7OtqMVNyJZDWM0iFPMQ0VxEIvaUCTiKomG1hrBZgthUH1atGVu0EVukATUaQtaT2zlLOJY0xgbiLTdRl5Wwy0rYoRK2a0QsMcJKmGiKdE0pKFg1C1bNii2iYg9o2PwRbC1hFC2MKZvg2iONXvPSnQSZdOH1Ql4e1NToXYluLFEfrqgvMTNqKw2ViNVL1OIlorqJKg6iOIhqdqIxO9GolVjMgqYp3bKAn6pqqGoEiyWCRQljUcKoSgiLFsSiBVG1AJaoD2ukCSWspe37UndQAKs/gtUfwbXdfTGrSsRpJWpTiVohZlOJWiBq0YgqGjE1RvSX/3UHFRUFBQsWLDEVS0zBElWwRDQsYQ1rKIbFH8YS0YgPdErRwU7dRbqV0oYEmXRSXp7WQWZHFGKJ7qld0VDQVBuaYiWmWNEUK5pqI4YFDSua+suvlBZDIYZCFEXTQIvGv//lv2jxP6uxIGosAjHkfchg1Ejsl0Xedi1qU9EsCpqqoimAqqApxL9UBZRtvldA0RTUmIbyy5ca+eXP0RhKJIaqAYkWIWlq6zCLBXr10rsK0UMkyKSTPn3g22/1riKlKWgovwyMTI0OKdETLOHYLy1lEjoMoVev+CQHkRZkTfh0snWVXyGEMDMZH5NWJMikG/kFF0KYnYyPSSsSZNKNBBkhhJnJar5pR4JMuikrA5ssGCeEMKmKCr0rED1Mgky6sVqhf3+9qxBCiK6nKDBokN5ViB4mQSYdDRmidwVCCNH1ysulWykNSZBJR/n5snS3EMJ85ENaWpIgk67kF14IYSZeb3wMoEg7EmTSVf/+YLfrXYUQQnSNIUPiY2RE2pEgk65k0K8QwixUVQb5pjEJMuls6FC9KxBCiM4rL5dVy9OYBJl0lpsb37ZACCFSmYz5S2sSZNKdtMoIIVJZZqYM8k1zEmTSXb9+MuhXCJG6Bg/WuwKhMwky6c5qhYED9a5CCCE6Tgb5CiTICJD+ZSFEaqqoAJdL7yqEziTICMjJgeJivasQQoiOkQ9hAgkyYqthw/SuQAgh2i8nB3r10rsKYQASZERcv37xFwYhhEgFY8boXYEwCAkyIk5RYOxYvasQQohdy8uLf/gSAgkyYlt9+8ZfIIQQwsikNUZsQ4KMSCatMkIIIysoiM9WEuIXEmREsvJy2bZACGFc8mFLbEeCjGhNXiiEEEZUVAS9e+tdhTAYCTKitbIyWVdGCGE848bpXYEwIAkyom3SKiOEMJLS0viXENuRICPaJi8aQggjkQ9XYgckyIgdk2ZcIYQRSHe32AkJMmLHZGCdEMIIpDVG7IQEGbFz8gIihNBTnz6yJITYKQkyYudk8SkhhJ6ki1vsggQZsWt77w1Wq95VCCHSzdChsm2K2CUJMmLXMjJkbxMhRM9yu2H8eL2rEClAgoxon5EjIT9f7yqEEOli4kSw2/WuQqQACTKifRQF9tsv/l8hhOhOFRXQt6/eVYgUIUFGtF9BAQwfrncVQggzs9nirTFCtJMEGdEx48aB16t3FUIIsxo/HjwevasQKUSCjOgYqxX23VfvKoQQZlRYGJ+pJEQHSJARHdenD1RW6l2FEMJMVBUmTZJxeKLDJMiI3bPPPuBw6F2FEMIs9tgDcnP1rkKkIAkyYve4XDBhgt5VCCHMICsL9txT7ypEipIgI3bf4MFQUqJ3FUKIVLfffmCx6F2FSFESZETnyAuQEKIzBg6E0lK9qxApTIKM6JzsbNm+QAixe9zu+F5uQnSCBBnReaNGQe/eelchhEglqgoHHSSTBkSnSZARXWPKFFkoTwjRfuPHQ3Gx3lUIE5AgI7qG0wkHHxz/lCWEEDvTt298I1ohuoC864iuU1AQX19GCCF2JCsLJk/WuwphIhJkRNcaOhT699e7CiGEEVmt8ZZbu13vSoSJSJARXW/SJMjJ0bsKIYTR7LuvrN4rupwEGdH1tn7qstn0rkQIYRSDB8fXjBGii0mQEd0jOzveMiOEEPn5MHGi3lUIk5IgI7pPZSUMH653FUIIPdnt8RZaWQFcdBMJMqJ77bUXFBbqXYUQQi9TpkBGht5VCBOTICO619bVO51OvSsRQvS0UaOgvFzvKoTJSZAR3c/rhQMPlMXyhEgnZWUwbpzeVYg0IO8somf06hVvYlYUvSsRQnS3wsL4uBj5fRc9QIKM6DmVlTJzQQizy8mBww6T5RdEj5EgI3rW0KEwdqzeVQghuoPXC4cfLjtaix4lQUb0vD33hBEj9K5CCNGVXC6YNg08Hr0rEWnGqncBIk3tvTcEg7Bkid6VpLWPlyzh9rff5ps1a9jY0MCL55/P0aNGJe6f9corPPPVV6ytq8NutTKmTx9uPvpoJvTtmzjm5tdf57UffmDB2rXYrVbq77476RqzP/2Usx5/vM3rb779dgozM/lk2TL+/MIL/LxpE75QiPLcXGZMmsSlBx2UVMsNr76a9PhBRUX8fOONnX8iROfY7fGWmKwsvSsRaUiCjNDP5MkQCsGqVXpXkrZaQiH2KCvj7IkTOfahh1rdP7CoiPtOOYV++fn4w2Huevddpt59N8tuuomCX9YGCUUinDBmDHv368ej8+a1OsdJY8dy6LBhSbdNnz2bQCRCYWYmAB67nQv335+RZWV47HY+WbaMGXPm4LHbOW+bFaKHlZby7iWXJL63yiJr+rNa4dBDIS9P70pEmpIgI/SjKPFp2W+8ARs26F1NWjps+HAO28nqy6eOH5/0/Z0nnMCj8+bx/bp1HDhkCAA3HHkkEG95aYvLbse1zW7HVU1NvL94MY+ecUbittF9+jC6T5/E9xX5+bwwfz5zly1LCjJWVaVYPvUbx9Z1ooqL9a5EpDEZIyP0ZbHAIYfE92IRhhaKRPjH3LlkuVzs0bv3bp/n359/jttu5/g999zhMfPXrOHTFSuYvN0mg0u3bKH0iivod801nPboo6yprd3tOkQX2H9/2CaACqEHaZER+rPZ4v3rL78M9fV6VyO28+r333PyI4/gC4UoycrinUsuId/r3e3zPTpvHqeOH5/USrNV2Z//TFVzM5FolFm/+Q3n7rtv4r4Jffsye/p0BhUVsbGhgRtefZX9br+dH2fOJENWju55EydC//56VyGEBBlhEE7nr2GmuVnvasQ2pgwaxIJrr6W6uZl/fvIJJ/7jH3xx5ZWJ8S0d8dny5fy0cSNPnHVWm/fPvfxymoNBPl+xgitffJH+BQWc8kv31rZdYCPLypjQty/lV13Ff77+mnO2CTyiB4wZA9uNexJCL9K1JIxj6xoULpfelYhteBwO+hcWsle/fjx6xhlYLZY2B/W2xyPz5jGqd2/G7GD/nb75+Yzo1Yvf7bcflx54ILO2m6W0rWy3m4FFRSyrqtqtWsRuGjEiHmSEMAgJMsJYsrPhqKNkt1wDi8ViBCORDj+uORCIt560c3XnmKbt9DrNgQDLq6ookcG/PWfPPeNLJwhhINK1JIwnMzMeZl57Derq9K7G1JoDgaQWjZXV1SxYu5Zcj4c8j4ebX3+dI/fYg5KsLKqbm7n/ww9ZX1/PCdt8Il9TW0ttSwtramuJxmIsWLsWgP4FBXi3Gbvy7NdfE4nFOH3ChFZ13P/BB/TJzWXwL7NfPl66lDveeYc/TpmSOOay55/nNyNHUp6by4aGBma+8goWVeUU2ZiwZ+yzD+xkhpsQelE0TdP0LkKINgWD8anZW7boXYlpfbh4MVPuvLPV7WfuvTcPnXYapz7yCF+sWkV1czN5Hg/jKiq49vDDGVdRkTh2+uzZPP7ZZ63O8cGf/sT+gwYlvt/nb3+jb34+c845p9Wx977/Pg/PncvK6mqsqkplQQG/228/Zuy3H+ovu6af/M9/8vHSpdS0tFDg9bJv//7cfPTRVBYUdMEzIXZIVeNrPg0YoHclQrRJgowwtkgE3n4b1q3TuxIh0o/FEl8nZgdjmoQwAgkywvhiMXj/fVixQu9KhEgfdnt8jaeSEr0rEWKnJMiI1KBp8Pnn8MMPelcihPl5PHDYYZCbq3clQuySBBmRWn78ET77LB5shBBdLy8vHmLcbr0rEaJdJMiI1LNqVbyraTemAAshdqJ37/iYGJtN70qEaDcJMiI1VVXBm2+C3693JUKYw5Ah8W0HVFleTKQWCTIidTU1xadny/5MQnTO+PEwapTeVQixWyTIiNQWDsPcubBsmd6VCJF6XC6YMgXKyvSuRIjdJkFGmMPixTBvnoybEaK9SkvhgANkUK9IeRJkhHnU1cG778q2BkLsjKLE90zac8/4n4VIcRJkhLlEIvGWmcWL9a5ECONxu+OtMKWlelciRJeRICPMaelS+OST+BgaIUR8HMyUKfFxMUKYiAQZYV719fDee1BTo3clQuhHVWHsWJmVJExLgowwt2g0vhLwokV6VyJEz/N44MADobhY70qE6DYSZER6WLECPv4YQiG9KxGiZ/TpA/vvD06n3pUI0a0kyIj00dgY72qqqtK7EiG6j6rGF7gbOVLvSoToERJkRHqJxeIbT37zjQwEFuZTWgr77gvZ2XpXIkSPkSAj0lNLS3zszIoVelciROe5XLDXXjBggN6VCNHjJMiI9LZuXXyadmOj3pUI0XGKEt/scfx4sNv1rkYIXUiQESIahQUL4l/RqN7VCNE+BQWw336Qn693JULoSoKMEFs1NsZbZ9at07sSIXbMbo+3wAwZIlsMCIEEGSFaW7EiPn6mpUXvSoRINmBAfCyMrM4rRIIEGSHaEg7HZzb9+GN8ppMQesrJic9GKinRuxIhDEeCjBA7U1sb34Ry40a9KxHpyGaL71I9YkR8fRghRCsSZIRoj/Xr4dtvJdCInmG3w/Dh8QDjcOhdjRCGJkFGiI7YuDHe5bRhg96VCDNyOOLhZfhwmU4tRDtJkBFid2zeHA80MsNJdAWnM76lwLBh8e4kIUS7SZARojO2bIl3Oa1Zo3clIhW5XLDHHjB0KFitelcjREqSICNEV6iujgeaVav0rkSkArc7HmCGDJEAI0QnSZARoivV1sYDjezhJNri9cYDzODBYLHoXY0QpiBBRojuUFcHP/wQDzShkN7VCL0VFMRbXwYOlGnUQnQxCTJCdKdIJB5mFi+WqdvpxumMr8Q7aBDk5updjRCmJUFGiJ7S2BgPNEuWyPYHZqUo0Lt3PLyUl0vrixA9QIKMED1N0+LTthcvjg8Oli0QUl9WVjy8DBwYH8grhOgxEmSE0FMgAMuWxUNNTY3e1YiOsFqhX7/4wN3iYr2rESJtSZARwiiqq+OBZvnyeMARxqMoUFQUb3mprJTF64QwAAkyQhhRVRWsXRvvgtq8Od4dJfThdkNZWXzsS1mZ7H0khMFIkBHC6EKheKDZGmxkoHD3UtV4q0vv3vGvvDy9KxJC7IQEGSFSTW1tPNSsXQubNslg4a7g9f7a6tKrl2zYKEQKkSAjRCoLh2H9+nhLzbp18SneYtes1uRWl5wcvSsSQuwmCTJCmEkwGJ/9tO1XXV16t9q43fEF6fLzf/1vVlZ84K4QIuVJkBHC7GKxeJjZPuAEg3pX1rUUBbKz42Natv1yufSuTAjRjSTICJGumpvj422qq+NBp6UF/H7w+eJdVkakqvGl/93ueEDJzPw1sOTmykaMQqQhCTJCiNYikXio2RpsfL7k77f9byTSuWspSjycuFy/BpRt/7ztf53Orvn5hBCmIUFGCNE5kUj8S9Pi3Vhb/7v1z5oWb0lRlOT/bv2z3S7jVYQQu02CjBBCCCFSlmzNKoQQQoiUJUFGCCGEEClLgowQQgghUpYEGSGEEEKkLAkyQgghhEhZEmSEEEIIkbIkyAghhBAiZUmQEUIIIUTKkiAjhBBCiJQlQUYIIYQQKUuCjBBCCCFSlgQZIYQQQqQsCTJCCCGESFkSZIQQQgiRsiTICCGEECJlSZARQgghRMqSICOE0M2HH36Ioih8+OGH3XJ+RVGYNWtWt5xbCGEMEmSESGGzZ89GURS+/vrrNu/ff//9GT58eA9XJYQQPceqdwFCCNFd/H4/Vqu8zAlhZvIbLoQwLafTqXcJQohuJl1LQqSZJ598kjFjxuByucjNzeXkk09m7dq1Scds7ZL6/vvvmTx5Mm63m/79+/P8888D8NFHHzFhwgRcLheDBg3i3XffbXWd+fPnc9hhh5GZmYnX6+XAAw/k888/32ltM2fOxGazUVVV1eq+8847j+zsbAKBAABff/01hxxyCPn5+bhcLvr27cvZZ5+d9Jjtx8g0NTVxySWXUFFRgcPhoLCwkIMPPphvv/22Xc+dEMJ4JMgIYQINDQ1UV1e3+gqHw0nH3XzzzZxxxhkMGDCAO++8k0suuYT33nuPSZMmUV9fn3RsXV0dRxxxBBMmTOC2227D4XBw8skn8+yzz3LyySdz+OGHc+utt9LS0sLxxx9PU1NT4rELFy5kv/3247vvvuOKK67guuuuY+XKley///588cUXO/w5fvvb3xKJRHj22WeTbg+FQjz//PMcd9xxOJ1OtmzZwtSpU/n/7dtfSFNtHAfw7zyWZJuboOHEQbZUZGBWsBlIuygMI3JpmDXbRUjk9NILvcqBQzSiC21hXoz+bHghSiRUGowoKN1VFGmIthuFGV2UIezCPe+FvIf3NF29vHtzJ78fOLD9znOe83vO1Y/fc04kEkFnZycGBgbgdDp/Wihdu3YNd+7cQUNDA3w+Hzo6OrBnzx7Mzs7+4pMmorQjiEi1/H6/AJD0sFgsQgghIpGIkCRJeL1exRzv3r0TmZmZirjdbhcARDAYlGNzc3MCgMjIyBBv3ryR48+ePRMAhN/vl2MOh0Ps3r1bLCwsyLHl5WWh0+nE8ePH5VgoFBIARCgUkmPHjh0TNptNkePY2Jhi3Pj4uAAgwuFw0ucDQFy/fl3+r9frRVtbW9JriEhd2JEh+gPcvn0bU1NTCUdFRYU8ZmxsDPF4HI2NjYquTUFBAUpKShAKhRRzarVaNDU1yf/LyspgMBhQXl4Om80mx//+vbi4CABYX1/H5OQkHA4HDhw4II8zGo24dOkSXr16hW/fvm25FpfLhenpaSwsLMixQCAAk8kEu90OADAYDACAiYmJhK5TMgaDAdPT01heXv7la4govbGQIfoDWK1WnDx5MuHIzc2Vx8zPz0MIgZKSEuTn5yuO2dlZrKysKOYsKiqCRqNRxPR6PUwmU0IM2NiKAoDPnz9jbW0NZWVlCXmWl5cjHo8nvJPzTxcuXEBWVhYCgQCAjW2ziYkJOJ1OOR+73Y6GhgZ4PB7k5eWhrq4Ofr8fsVgs6XPq7+/H+/fvYTKZYLVa0d3dLRdgRKROLGSIdoh4PA6NRoOnT59u2r0ZGhpSjJckadN5tooLIVKSZ25uLs6cOSMXMqOjo4jFYmhubpbHaDQajI6O4vXr12hvb8fS0hKuXLmCo0eP4vv371vO3djYiMXFRQwMDKCwsBA3btyAxWLBkydPUpI7Ef1+/PyaaIcwm80QQqC4uBilpaX/233y8/ORnZ2Njx8/Jpybm5tDRkZGQlfnRy6XC3V1dQiHwwgEAjh8+DAsFkvCuKqqKlRVVcHr9SIYDMLpdGJkZAQtLS1bzm00GuF2u+F2u7GysoIjR47A6/Witrb23y+WiLYdOzJEO0R9fT0kSYLH40nonggh8OXLl5TcR5Ik1NTU4NGjR4hEInI8Go0iGAyiuroaOTk5Seeora1FXl4e+vr68OLFC0U3BtjYxvpxDZWVlQCw5fbS+vo6vn79qojt27cPhYWFP92SIqL0xY4M0Q5hNpvR09ODrq4uRCIROBwO6HQ6fPr0CePj47h69So6OjpScq+enh5MTU2huroabrcbmZmZGBoaQiwWQ39//0+v37VrF5qamjA4OAhJknDx4kXF+Xv37sHn8+HcuXMwm81YXV3F8PAwcnJycPr06U3nXF1dRVFREc6fP49Dhw5Bq9Xi+fPnCIfDuHnzZkrWTUS/HwsZoh2ks7MTpaWluHXrFjweDwDAZDKhpqYGZ8+eTdl9LBYLXr58ia6uLvT29iIej8Nms+Hhw4eKL56ScblcGBwcxIkTJ2A0GhXn7HY7ZmZmMDIygmg0Cr1eD6vVikAggOLi4k3ny87OhtvtxuTkpPwF18GDB+Hz+dDa2vqf10xE20MjUvWGHhFRCr19+xaVlZW4f/8+Ll++vN3pEFGa4jsyRJSWhoeHodVqUV9fv92pEFEa49YSEaWVx48f48OHD7h79y7a29uxd+/e7U6JiNIYt5aIKK3s378f0WgUp06dwoMHD6DT6bY7JSJKYyxkiIiISLX4jgwRERGpFgsZIiIiUi0WMkRERKRaLGSIiIhItVjIEBERkWqxkCEiIiLVYiFDREREqsVChoiIiFSLhQwRERGpFgsZIiIiUi0WMkRERKRaLGSIiIhItf4Cg7+jzyaElUIAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 700x700 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Define the sets as lists of elements\n",
    "set1 = set(data.query('MBC_8_prob == 100')['ID'].values)\n",
    "set2 = set(data.query('selectivity_pred < 0.5')['ID'].values)\n",
    "set3 = set(data.query('hemolysis_pred <= 2')['ID'].values)\n",
    "\n",
    "\n",
    "# Create the Venn diagram\n",
    "plt.figure(figsize=(7,7))\n",
    "plt.title('Bacterially Specific')\n",
    "venn3_unweighted([set1, set2, set3], set_labels=('Activity', 'Selectivity', 'Hemolysis'), set_colors=('b', 'g', 'r'))\n",
    "#plt.savefig('results/venn3_predictions_all_mutations_bac_mamm_specific.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 700x700 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Define the sets as lists of elements\n",
    "set1 = set(data.query('MBC_8_prob == 0')['ID'].values)\n",
    "set2 = set(data.query('selectivity_pred > 2.35')['ID'].values)\n",
    "set3 = set(data.query('hemolysis_pred > 50')['ID'].values)\n",
    "\n",
    "\n",
    "# Create the Venn diagram\n",
    "plt.figure(figsize=(7,7))\n",
    "plt.title('Mammalian Specific')\n",
    "venn3_unweighted([set1, set2, set3], set_labels=('Activity', 'Selectivity', 'Hemolysis'), set_colors=('b', 'g', 'r'))\n",
    "\n",
    "#plt.savefig('results/venn3_predictions_all_mutations_mammalian_specific.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# AA Frequency Change "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of Bacterial specific peptides: 95472\n",
      "Number of Mamalian specific peptides: 51876\n",
      "Number of Most selective peptides: 118685\n",
      "Number of Least hemolitic peptides: 918364\n",
      "Number of Most active peptides: 3443109\n"
     ]
    }
   ],
   "source": [
    "### Bacterial and Mammalian specific peptides\n",
    "bac = data.query('MBC_8_prob == 100 and selectivity_pred < 0.5 and hemolysis_pred <=2').copy()\n",
    "\n",
    "### Mamalian specific peptides\n",
    "mamm = data.query('MBC_8_prob == 0 and selectivity_pred > 2.35 and hemolysis_pred > 50').copy()\n",
    "\n",
    "### Most selective\n",
    "most_sel = data.query('selectivity_pred < 0.5').copy()\n",
    "\n",
    "### Least hemolitic\n",
    "least_hemo = data.query('hemolysis_pred <= 2').copy()\n",
    "\n",
    "### Most active\n",
    "most_act = data.query('MBC_8_prob == 100').copy()\n",
    "\n",
    "\n",
    "print(f\"Number of Bacterial specific peptides: {bac.shape[0]}\")\n",
    "print(f\"Number of Mamalian specific peptides: {mamm.shape[0]}\")\n",
    "print(f\"Number of Most selective peptides: {most_sel.shape[0]}\")\n",
    "print(f\"Number of Least hemolitic peptides: {least_hemo.shape[0]}\")\n",
    "print(f\"Number of Most active peptides: {most_act.shape[0]}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'-RR-I---PL--Y-----'"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# def to find the differences between the residues\n",
    "idx = \"RRRRICYCPLRFYVCVGR\"\n",
    "\n",
    "def diff_func(idxx):\n",
    "    pg1 = \"RGGRLCYCRRRFCVCVGR\"\n",
    "    idx = idxx\n",
    "    diff = []\n",
    "    for i in range(0, len(pg1)):\n",
    "        aa = idx[i]\n",
    "        pg_aa = pg1[i]\n",
    "        \n",
    "        if pg_aa == aa:\n",
    "            diff.append(\"-\")\n",
    "        else:\n",
    "            diff.append(aa)\n",
    "    diff_seq = \"\".join(diff)\n",
    "    return diff_seq\n",
    "\n",
    "\n",
    "diff_func(idx)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# HeatMaps"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Most Active Peptides"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# prepare the matrix to plot the heatmap\n",
    "most_act[\"difference_in_seq\"] = [diff_func(x) for x in most_act[\"Sequence\"]]\n",
    "diff = most_act[\"difference_in_seq\"].str.split(\"\", n= -1, expand = True)\n",
    "most_act = most_act.merge(diff, left_index=True, right_index=True)\n",
    "\n",
    "\n",
    "# charged->polar->aliphatic->aromatic->other\n",
    "aa_list = ['H', 'K', 'R', 'D', 'E', 'N', 'Q', 'S', 'T', 'A', 'I', 'L', 'V', 'M', 'F', 'W', 'Y','G', 'C', 'P']\n",
    "\n",
    "#aa_list = ['H', 'K', 'R', 'D', 'E', 'A', 'I', 'L', 'V', 'M', 'F', 'W', 'Y', 'N', 'Q', 'S', 'T', 'G', 'C', 'P']\n",
    "mtx = pd.DataFrame(0, index=aa_list, columns=np.arange(1, 19))\n",
    "\n",
    "# fill the matix with the count of each amino acid in each position\n",
    "for i in range(1, 19):\n",
    "    pos_count = most_act[i].value_counts().to_dict()\n",
    "\n",
    "    for k,v in pos_count.items():\n",
    "        if k != '-':\n",
    "            mtx.loc[k,i] = v\n",
    "\n",
    "# plot the heatmap\n",
    "plt.title(\"Most Active Peptides\")\n",
    "sns.heatmap(mtx.T, cmap=\"rocket_r\")\n",
    "#plt.savefig('results/changes_heatmap_predictions_most_active.png', dpi=1200)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Least Hemolitic Peptides"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# prepare the matrix to plot the heatmap\n",
    "least_hemo[\"difference_in_seq\"] = [diff_func(x) for x in least_hemo[\"Sequence\"]]\n",
    "diff = least_hemo[\"difference_in_seq\"].str.split(\"\", n= -1, expand = True)\n",
    "least_hemo =least_hemo.merge(diff, left_index=True, right_index=True)\n",
    "\n",
    "\n",
    "# charged->polar->aliphatic->aromatic->other\n",
    "aa_list = ['H', 'K', 'R', 'D', 'E', 'N', 'Q', 'S', 'T', 'A', 'I', 'L', 'V', 'M', 'F', 'W', 'Y','G', 'C', 'P']\n",
    "\n",
    "#aa_list = ['H', 'K', 'R', 'D', 'E', 'A', 'I', 'L', 'V', 'M', 'F', 'W', 'Y', 'N', 'Q', 'S', 'T', 'G', 'C', 'P']\n",
    "mtx = pd.DataFrame(0, index=aa_list, columns=np.arange(1, 19))\n",
    "\n",
    "# fill the matix with the count of each amino acid in each position\n",
    "for i in range(1, 19):\n",
    "    pos_count = least_hemo[i].value_counts().to_dict()\n",
    "\n",
    "    for k,v in pos_count.items():\n",
    "        if k != '-':\n",
    "            mtx.loc[k,i] = v\n",
    "\n",
    "# plot the heatmap\n",
    "plt.title(\"Least Hemolitic Peptides\")\n",
    "sns.heatmap(mtx.T, cmap=\"rocket_r\")\n",
    "#plt.savefig('results/changes_heatmap_predictions_least_hemolitic.png' , dpi=1200)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Most Selective"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhAAAAGzCAYAAAB+YC5UAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/OQEPoAAAACXBIWXMAAA9hAAAPYQGoP6dpAABY8klEQVR4nO3deVhU1f8H8PcdlgFZXVlccMHcU8MlckEUwTUxl0zKXcstlcLCzCVLRMvdNL+mUrmXGlmiqJAbbqi5ZISKSypuKCjqsMz9/eGPyRmW4Q4HGen96rnPk/feec+ZK8iHc849V5JlWQYRERGRAqqSbgARERG9eFhAEBERkWIsIIiIiEgxFhBERESkGAsIIiIiUowFBBERESnGAoKIiIgUYwFBREREirGAICIiIsVYQBAZMWjQIFSvXv25v++lS5cgSRJWr1793N/b3KxevRqSJOHSpUtGz61evToGDRpU7G0i+q9jAUH5yvlHW5Ik7N+/P9dxWZZRtWpVSJKEbt26FUsbrl+/jmnTpuHkyZOFfs3p06fRu3dveHh4wMbGBpUrV0bHjh2xaNGiYmljUa1duxbz588v6Wbo5BQuOZuFhQWqVauGnj17Kvp7MMXMmTOxdevWYn0PIhKDBQQZZWNjg7Vr1+ba//vvv+Off/6BWq0utve+fv06pk+fXugfXAcPHkSzZs3wxx9/YPjw4Vi8eDGGDRsGlUqFBQsWFFs7iyK/AsLDwwOPHz/GO++88/wbBeCtt97C999/j5UrV6J///7Ys2cPXn311WItIvIrIN555x08fvwYHh4exfbeRKSMZUk3gMxfly5dsGnTJixcuBCWlv9+yaxduxZeXl64c+dOCbZO3xdffAEnJyccPXoUzs7Oesdu3bpVMo0ykSRJsLGxKbH3f+WVV/D222/r/tyqVSu8/vrrWLp0Kb755pvn2hYLCwtYWFg81/ckooKxB4KMeuutt3D37l1ER0fr9mVkZODHH39E//7983xNeno6PvjgA1StWhVqtRp16tTBl19+CcOHv0ZHR6N169ZwdnaGvb096tSpg0mTJgEAYmNj0bx5cwDA4MGDdV3qBc0JuHDhAho0aJCreACASpUq5dr3ww8/wMvLC7a2tihXrhz69euHq1evGrsk0Gq1mD9/Pho0aAAbGxu4uLjg3Xffxb1793Kdu337dvj4+MDBwQGOjo5o3ry5rkenXbt2+PXXX3H58mXd58uZb2E4B+LLL7+EJEm4fPlyrvcIDQ2FtbW13vsfPnwYnTp1gpOTE8qUKQMfHx8cOHDA6GfLT/v27QEASUlJit5j2rRpkCQJf/31F/r27QtHR0eUL18e48aNw5MnT3TnSZKE9PR0RERE6K5FzlyGvOZAyLKMzz//HFWqVEGZMmXg6+uLs2fP5tn2+/fvY/z48bqvR09PT4SHh0Or1eqdt379enh5een+rho1amS2PVdEJY0FBBlVvXp1eHt7Y926dbp927dvR2pqKvr165frfFmW8frrr2PevHno1KkT5s6dizp16iAkJATBwcG6886ePYtu3bpBo9Hgs88+w1dffYXXX39d9wOoXr16+OyzzwAAI0aMwPfff4/vv/8ebdu2zbetHh4eiI+Px5kzZ4x+ri+++AIDBgxA7dq1MXfuXIwfPx67d+9G27Ztcf/+/QJf++677yIkJAStWrXCggULMHjwYKxZswYBAQHIzMzUnbd69Wp07doVKSkpCA0NxaxZs9CkSRNERUUBAD755BM0adIEFSpU0H2+/OZD9O3bF5IkYePGjbmObdy4Ef7+/ihbtiwAYM+ePWjbti3S0tIwdepUzJw5E/fv30f79u1x5MgRo9cmLxcuXAAAlC9f3qT36Nu3L548eYKwsDB06dIFCxcuxIgRI3THv//+e6jVarRp00Z3Ld5999182zNlyhR8+umnaNy4MebMmYOaNWvC398f6enpeuc9evQIPj4++OGHHzBgwAAsXLgQrVq1QmhoqN7XY3R0NN566y2ULVsW4eHhmDVrFtq1a1ekoouoVJOJ8rFq1SoZgHz06FF58eLFsoODg/zo0SNZlmW5T58+sq+vryzLsuzh4SF37dpV97qtW7fKAOTPP/9cL693796yJEny+fPnZVmW5Xnz5skA5Nu3b+fbhqNHj8oA5FWrVhWqzTt37pQtLCxkCwsL2dvbW544caK8Y8cOOSMjQ++8S5cuyRYWFvIXX3yht//06dOypaWl3v6BAwfKHh4euj/v27dPBiCvWbNG77VRUVF6++/fvy87ODjILVu2lB8/fqx3rlar1f1/165d9fJzJCUl5frs3t7espeXl955R44ckQHI3333nS67du3ackBAgN77PHr0SK5Ro4bcsWPHXO+V1/tOnz5dvn37tpycnCzHxsbKTZs2lQHIP/30k6L3mDp1qgxAfv311/XeZ9SoUTIA+Y8//tDts7OzkwcOHJirTTlfi0lJSbIsy/KtW7dka2truWvXrnrvP2nSJBmAXsaMGTNkOzs7+e+//9bL/Pjjj2ULCwv5ypUrsizL8rhx42RHR0c5KyurwOtDRE+xB4IKpW/fvnj8+DG2bduGBw8eYNu2bfkOX/z222+wsLDA+++/r7f/gw8+gCzL2L59OwDohhl+/vnnXF3JpurYsSPi4uLw+uuv448//sDs2bMREBCAypUrIzIyUnfe5s2bodVq0bdvX9y5c0e3ubq6onbt2oiJicn3PTZt2gQnJyd07NhR77VeXl6wt7fXvTY6OhoPHjzAxx9/nGsugyRJJn2+N998E/Hx8breAADYsGED1Go1evToAQA4efIkEhMT0b9/f9y9e1fXvvT0dHTo0AF79+4t1PWeOnUqKlasCFdXV7Rr1w4XLlxAeHg43njjDZPeY/To0Xp/Hjt2LICnXy9K7dq1CxkZGRg7dqzetRw/fnyuczdt2oQ2bdqgbNmyen9ffn5+yM7Oxt69ewE8/XpMT0/XG6ojovxxEiUVSsWKFeHn54e1a9fi0aNHyM7ORu/evfM89/Lly3B3d4eDg4Pe/nr16umOA09/GK5YsQLDhg3Dxx9/jA4dOuCNN95A7969oVKZXts2b94cmzdvRkZGBv744w9s2bIF8+bNQ+/evXHy5EnUr18fiYmJkGUZtWvXzjPDysoq3/zExESkpqbmOacC+HeyZs4P+YYNG5r8WQz16dMHwcHB2LBhAyZNmgRZlrFp0yZ07twZjo6OuvYBwMCBA/PNSU1N1Q135GfEiBHo06cPVCoVnJ2d0aBBA90dN6a8h+G1rlWrFlQqVaHWdjCU8zVkmFmxYsVcnysxMRGnTp1CxYoV88zK+fsaNWoUNm7ciM6dO6Ny5crw9/dH37590alTJ8XtI/ovYAFBhda/f38MHz4cycnJ6Ny5c54TFZWwtbXF3r17ERMTg19//RVRUVHYsGED2rdvj507dxZ51r21tTWaN2+O5s2b46WXXsLgwYOxadMmTJ06FVqtFpIkYfv27Xm+j729fb65Wq0WlSpVwpo1a/I8nt8PKhHc3d3Rpk0bbNy4EZMmTcKhQ4dw5coVhIeH67UPAObMmYMmTZrkmVPQ58tRu3Zt+Pn55XlMxHuY2gujlFarRceOHTFx4sQ8j7/00ksAnk6yPXnyJHbs2IHt27dj+/btWLVqFQYMGICIiIjn0laiFwkLCCq0nj174t1338WhQ4ewYcOGfM/z8PDArl278ODBA71eiL/++kt3PIdKpUKHDh3QoUMHzJ07FzNnzsQnn3yCmJgY+Pn5Cfsh06xZMwDAjRs3ADz97VeWZdSoUUP3A6SwatWqhV27dqFVq1awtbUt8DwAOHPmDDw9PfM9T+lnfPPNNzFq1CgkJCRgw4YNKFOmDLp3757rfR0dHfMtAIrKlPdITExEjRo1dH8+f/48tFqt3iqfhb0WOV9DiYmJqFmzpm7/7du3c90JU6tWLTx8+LBQ7bS2tkb37t3RvXt3aLVajBo1Ct988w0+/fTTAv8Oif6LOAeCCs3e3h5Lly7FtGnT9H5gGerSpQuys7OxePFivf3z5s2DJEno3LkzACAlJSXXa3N+m9VoNAAAOzs7ADB6V0SOmJiYXLeKAv+Os9epUwcA8MYbb8DCwgLTp0/Pdb4sy7h7926+79G3b19kZ2djxowZuY5lZWXp2urv7w8HBweEhYXp3a6Y8x457OzskJqaWqjPBwC9evWChYUF1q1bh02bNqFbt2666wQAXl5eqFWrFr788ks8fPgw1+tv375d6PfKjynvsWTJEr0/56wMmvP1ADy9FoX5u/bz84OVlRUWLVqkdy3zuoOlb9++iIuLw44dO3Idu3//PrKysgAg19+5SqXCyy+/DODfr0ci+hd7IEiRgsa8c3Tv3h2+vr745JNPcOnSJTRu3Bg7d+7Ezz//jPHjx+t+e/3ss8+wd+9edO3aFR4eHrh16xa+/vprVKlSBa1btwbw9LdHZ2dnLFu2DA4ODrCzs0PLli31fpN91tixY/Ho0SP07NkTdevWRUZGBg4ePIgNGzagevXqGDx4sC73888/R2hoKC5duoTAwEA4ODggKSkJW7ZswYgRI/Dhhx/m+R4+Pj549913ERYWhpMnT8Lf3x9WVlZITEzEpk2bsGDBAvTu3RuOjo6YN28ehg0bhubNm6N///4oW7Ys/vjjDzx69EjXLe7l5YUNGzYgODgYzZs3h729fYEFWqVKleDr64u5c+fiwYMHePPNN/WOq1QqrFixAp07d0aDBg0wePBgVK5cGdeuXUNMTAwcHR3xyy+/GP17LIgp75GUlITXX38dnTp1QlxcHH744Qf0798fjRs31p3j5eWFXbt2Ye7cuXB3d0eNGjXQsmXLXO9fsWJFfPjhhwgLC0O3bt3QpUsXnDhxAtu3b0eFChX0zg0JCUFkZCS6deuGQYMGwcvLC+np6Th9+jR+/PFHXLp0CRUqVMCwYcOQkpKC9u3bo0qVKrh8+TIWLVqEJk2a6ObvENEzSuz+DzJ7z97GWRDD2zhlWZYfPHggT5gwQXZ3d5etrKzk2rVry3PmzNG75W737t1yjx49ZHd3d9na2lp2d3eX33rrrVy32/38889y/fr1ZUtLS6O3dG7fvl0eMmSIXLduXdne3l62traWPT095bFjx8o3b97Mdf5PP/0kt27dWrazs5Pt7OzkunXryqNHj5YTEhJ05xjexplj+fLlspeXl2xrays7ODjIjRo1kidOnChfv35d77zIyEj5tddek21tbWVHR0e5RYsW8rp163THHz58KPfv3192dnaWAejeK6/bOHP873//kwHIDg4OuW4RzXHixAn5jTfekMuXLy+r1WrZw8ND7tu3r7x79+58r9+z7ztnzpwCzyvse+Tcxvnnn3/KvXv3lh0cHOSyZcvKY8aMydX2v/76S27btq1sa2urdzum4W2csizL2dnZ8vTp02U3NzfZ1tZWbteunXzmzBnZw8Mj162gDx48kENDQ2VPT0/Z2tparlChgvzaa6/JX375pe4W3x9//FH29/eXK1WqJFtbW8vVqlWT3333XfnGjRtGrwPRf5Eky3n09xIRCTJt2jRMnz4dt2/fztU7QEQvLs6BICIiIsVYQBAREZFiLCCIiIhIMc6BICIiIsXYA0FERESKsYAgIiIixVhAEBERkWJmsxJl5o1zYgMtBH40VdEe6mRIfph7Cecisc7/eQwmEfRo7RyyJl1onqqMk7As7cN7xk9SQLIT1zYAkCytheaJ/lqmItBml3QLCmbm/+5ZV28mNM+QpXVlYVlZGdeEZZkTsykgiIiIzMXzeVbsi41DGERERKQYeyCIiIgMFPbR8v9lLCCIiIgMsIAwjgUEERGRAYmzIIziHAgiIiJSjD0QREREBjiEYRwLCCIiIgMqFhBGCR/CuHr1KoYMGVLgORqNBmlpaXqbRpMhuilERERUTIQXECkpKYiIiCjwnLCwMDg5Oelt4YuWi24KERGRSSSB/5VWiocwIiMjCzx+8eJFoxmhoaEIDg7W26dKSVLaFCIiomLBIQzjFBcQgYGBkCQJsizne46xySdqtRpqtVpvX2a64DX/iYiIqNgoHsJwc3PD5s2bodVq89yOHz9eHO0kIiJ6biRJEraVVooLCC8vL8THx+d73FjvBBERkblTQRK2lVaKhzBCQkKQnp7/45k9PT0RExNTpEYRERGReVNcQLRp06bA43Z2dvDx8TG5QURERCWtNA89iMKFpIiIiAzwLgzjWEAQEREZYA+EcXyYFhERESlmNj0QcnaW0DxJJa420iZfEpYFANrbl4XmWXg2F5qnTbkmNA8P7wmNk6s1FJalTTgkLAsALOq3FpoHO9HfohZi47IzxeaJZGFV0i14vlSC/24Fk9R2Jd0ERUrz3ROimE0BQUREZC44hGEchzCIiIhIMfZAEBERGSjND8EShQUEERGRAd7GaRyHMIiIiEgx9kAQEREZ4CRK41hAEBERGeBtnMZxCIOIiIgUK5EeCI1GA41Go7dP0mRArbYuieYQERHp4RCGcYp7IB4/foz9+/fjzz//zHXsyZMn+O6774xmhIWFwcnJSW+bvWSF0qYQEREVCxUkYVtppaiA+Pvvv1GvXj20bdsWjRo1go+PD27cuKE7npqaisGDBxvNCQ0NRWpqqt42cfQw5a0nIiIqBpKkEraVVoo+2UcffYSGDRvi1q1bSEhIgIODA1q1aoUrV64oelO1Wg1HR0e9jcMXRERELw5FcyAOHjyIXbt2oUKFCqhQoQJ++eUXjBo1Cm3atEFMTAzs7F6sh6UQERHlhStRGqeoB+Lx48ewtPy35pAkCUuXLkX37t3h4+ODv//+W3gDiYiInjeVJAnbSitFPRB169bFsWPHUK9ePb39ixcvBgC8/vrr4lpGREREZktRD0TPnj2xbt26PI8tXrwYb731FmRZFtIwIiKikiIJ/K+0UlRAhIaG4rfffsv3+Ndffw2tVlvkRhEREZUkDmEYV3rvLyEiIqJiw2dhEBERGSjNQw+isIAgIiIyUJqHHkQxmwJCsrYRmidnPBGX9eCOsCwAUFVrKDQPFmL/GiVLtdA82dZBaJ7IrxWprIuwLACQNelC8yTHCkLzhLOwKukWvLhUFiXdgudLxRHz0sZsCggiIiJzwSEM41hAEBERGeAQhnEsIIiIiAywB8I4DkoRERGZiezsbHz66aeoUaMGbG1tUatWLcyYMUNvkUZZljFlyhS4ubnB1tYWfn5+SExM1MtJSUlBUFAQHB0d4ezsjKFDh+Lhw4d655w6dQpt2rSBjY0NqlatitmzZytqKwsIIiIiAypIwjYlwsPDsXTpUixevBjnzp1DeHg4Zs+ejUWLFunOmT17NhYuXIhly5bh8OHDsLOzQ0BAAJ48+ffmgaCgIJw9exbR0dHYtm0b9u7dixEjRuiOp6Wlwd/fHx4eHoiPj8ecOXMwbdo0LF++vNBt5RAGERGRAamE5kAcPHgQPXr0QNeuXQEA1atXx7p163DkyBEAT3sf5s+fj8mTJ6NHjx4AgO+++w4uLi7YunUr+vXrh3PnziEqKgpHjx5Fs2bNAACLFi1Cly5d8OWXX8Ld3R1r1qxBRkYGVq5cCWtrazRo0AAnT57E3Llz9QqNgrAHgoiIqBhpNBqkpaXpbRqNJs9zX3vtNezevVv3dOs//vgD+/fvR+fOnQEASUlJSE5Ohp+fn+41Tk5OaNmyJeLi4gAAcXFxcHZ21hUPAODn5weVSoXDhw/rzmnbti2sra115wQEBCAhIQH37t0r1OdiAUFERGRA5BBGWFgYnJyc9LawsLA83/fjjz9Gv379ULduXVhZWaFp06YYP348goKCAADJyckAABcX/TVsXFxcdMeSk5NRqVIlveOWlpYoV66c3jl5ZTz7HsYoHsI4d+4cDh06BG9vb9StWxd//fUXFixYAI1Gg7fffhvt27c3mqHRaHJVXypNBtRq63xeQURE9PyIHMIIDQ1FcHCw3j61Ou8F+zZu3Ig1a9Zg7dq1umGF8ePHw93dHQMHDhTWJhEU9UBERUWhSZMm+PDDD9G0aVNERUWhbdu2OH/+PC5fvgx/f3/s2bPHaE5e1Vj4wmUmfwgiIiJzpVar4ejoqLflV0CEhIToeiEaNWqEd955BxMmTND1WLi6ugIAbt68qfe6mzdv6o65urri1q1besezsrKQkpKid05eGc++hzGKCojPPvsMISEhuHv3LlatWoX+/ftj+PDhiI6Oxu7duxESEoJZs2YZzQkNDUVqaqre9tH77ylpChERUbGRBP6nxKNHj6AyWPbbwsICWq0WAFCjRg24urpi9+7duuNpaWk4fPgwvL29AQDe3t64f/8+4uPjdefs2bMHWq0WLVu21J2zd+9eZGZm6s6Jjo5GnTp1ULZs2UK1VVEBcfbsWQwaNAgA0LdvXzx48AC9e/fWHQ8KCsKpU6eM5uRdjXH4goiIzENJ3cbZvXt3fPHFF/j1119x6dIlbNmyBXPnzkXPnj0BPB1aGT9+PD7//HNERkbi9OnTGDBgANzd3REYGAgAqFevHjp16oThw4fjyJEjOHDgAMaMGYN+/frB3d0dANC/f39YW1tj6NChOHv2LDZs2IAFCxbkGmopiOI5EDnjQiqVCjY2NnByctIdc3BwQGpqqtJIIiIiwtPbLT/99FOMGjUKt27dgru7O959911MmTJFd87EiRORnp6OESNG4P79+2jdujWioqJgY/PvgwbXrFmDMWPGoEOHDlCpVOjVqxcWLlyoO+7k5ISdO3di9OjR8PLyQoUKFTBlypRC38IJAJL87PJWRjRu3Bjh4eHo1KkTAODMmTOoW7cuLC2f1iH79u3DwIEDcfHixUI3IEfmrUTjJykg8mmc2qtnhGUBgFSphtg8wU+7lO8VbgZuofMEP6HSonIdYVnZ548JywIAqZy70DxVRQ+heUQlJjvT+DkKWLmI+3cgL96VfYVlxV2LEZZlThT1QIwcORLZ2dm6PzdsqP9Y6u3btxfqLgwiIiJzpnTo4b9IUQHx3nsFT3ScOXNmkRpDRERkDlhAGMeFpIiIiEgxPguDiIjIAPsfjGMBQUREZEBVQg/TepGYTwGhshAaJ1mJW1fConZLYVkAAG228XOUEH3tKlQRmic/fiA0DxZW4rLsnIyfo4DKuXAruBERvejMp4AgIiIyE0pXkPwvYgFBRERkgHdhGMe7MIiIiEgx9kAQEREZEPk479KKBQQREZEBDmEYxyEMIiIiUow9EERERAZ4F4ZxQgoIWZY5XkRERKUGu+eNE3KN1Go1zp07JyKKiIioxEmSJGwrrRT1QAQHB+e5Pzs7G7NmzUL58uUBAHPnzi0wR6PRQKPR6O1TaTRQq9VKmkNEREQlRFEBMX/+fDRu3BjOzs56+2VZxrlz52BnZ1eoaissLAzTp0/X2zc55H1MmThOSXOIiIiKBe/CME6SZVku7MmzZs3C8uXLsWLFCrRv316338rKCn/88Qfq169fqJw8eyAeXBPbA5GdKS5L5LMXALN/FobQawfxz8KQ7MsJy8r+509hWQBg4VJLaB6s2CtHpYTgf1esXOoIzTPUvVo3YVm/XNkmLMucKJoD8fHHH2PDhg0YOXIkPvzwQ2RmmvYFoVar4ejoqLdx+IKIiOjFoXgSZfPmzREfH4/bt2+jWbNmOHPmTKmeJEJERP89KoFbaWXSbZz29vaIiIjA+vXr4efnh+xswV3yREREJYhzIIwr0joQ/fr1Q+vWrREfHw8PDw9RbSIiIiIzV+SFpKpUqYIqVaqIaAsREZFZ4NC8cVzKmoiIyEBpnrsgCq8RERERKcYeCCIiIgN8mJZxZlNAaFOuC82TbB2EZWmvHBOWVRwkp0pC8+R7yULzsmOihOZZdH5DWFb6tIXCsgCgzJheQvNUtV4Rm1fOXWie6EXC5LQ7wrKksq7CsgBAsrYVmidnPBaaB9F5gmX9vEJontW4ZULzDPEuDOPMpoAgIiIyFxzfN47XiIiIiBRjDwQREZEBzoEwjgUEERGRAc6BMI5DGERERKQYeyCIiIgMsP/BOBYQREREBlRcytooDmEQERGRYkXqgUhPT8fGjRtx/vx5uLm54a233kL58uWNvk6j0UCj0ejvzMiA2tq6KM0hIiISgr9dG6foGtWvXx8pKSkAgKtXr6Jhw4aYMGECoqOjMXXqVNSvXx9JSUlGc8LCwuDk5KS3zf7mB9M+ARERkWCSwP9KK0UFxF9//YWsrCwAQGhoKNzd3XH58mUcOXIEly9fxssvv4xPPvnEaE5oaChSU1P1tonvvm3aJyAiIqLnzuQhjLi4OCxbtgxOTk4AAHt7e0yfPh39+vUz+lq1Wg21Wq23T8PhCyIiMhMcwjBOcQEh/f/M1CdPnsDNzU3vWOXKlXH79m0xLSMiIiohpXnoQRTFBUSHDh1gaWmJtLQ0JCQkoGHDhrpjly9fLtQkSiIiInPGHgjjFBUQU6dO1fuzvb293p9/+eUXtGnTpuitIiIiIrNWpALC0Jw5c4rUGCIiInPAZ2EYx5UoiYiIDLB8MI7DPERERKQYeyCIiIgMcAjDOLMpILL3bBKaJ9+9JyzrUexlYVkAcPOCg9C8SjUeCM2zKCM0DiePuArNa5a0VFjWuDNi7xr6dMJ2oXnVZmcIzdM6VxCahwt/is2r5C4u67bY71sLz+ZC87IPbhGahwyN8XMUsOw0VGje413nhOaVGSc0Lhd2zxvHa0RERESKmU0PBBERkbngQlLGsYAgIiIywO5543iNiIiISDH2QBARERngAIZxLCCIiIgM8DZO41hAEBERGeD4vnG8RkRERKQYeyCIiIgMcADDOEU9EMePH0dSUpLuz99//z1atWqFqlWronXr1li/fn2hcjQaDdLS0vQ2TWaWspYTEREVExUkYVtppaiAGDx4MC5cuAAAWLFiBd599100a9YMn3zyCZo3b47hw4dj5cqVRnPCwsLg5OSkt32545hpn4CIiIieO0VDGImJiahduzYA4Ouvv8aCBQswfPhw3fHmzZvjiy++wJAhQwrMCQ0NRXBwsN6+7IhQJU0hIiIqNpwgaJyiAqJMmTK4c+cOPDw8cO3aNbRo0ULveMuWLfWGOPKjVquhVqv19j2y4nQMIiIyD6V34EEcRUVW586dsXTp0ych+vj44Mcff9Q7vnHjRnh6eoprHREREZklRb/2h4eHo1WrVvDx8UGzZs3w1VdfITY2FvXq1UNCQgIOHTqELVsEP6KWiIjoOSvNkx9FUdQD4e7ujhMnTsDb2xtRUVGQZRlHjhzBzp07UaVKFRw4cABdunQprrYSERE9FypZ3FZaKZ544OzsjFmzZmHWrFnF0R4iIiJ6AXDmIhERkQHehWEcCwgiIiIDnAFhHAsIIiIiA+yBMM5sCgiLdm8IzZPvJwvLslNHC8sCgJqODkLzVI1aGD9JAfnSX0LzvFtfEZqneq2bsKxhv0UKywKAav/rLjTPol4roXmStY3QPNTxFhqnfXhPWJaqjJOwLACAldr4OQpYtn9baJ6cdkdoHrIzhcaVea+n0DwqeWZTQBAREZkL3sZpHHtpiIiIDEgCN6WuXbuGt99+G+XLl4etrS0aNWqEY8f+fV6ULMuYMmUK3NzcYGtrCz8/PyQmJuplpKSkICgoCI6OjnB2dsbQoUPx8OFDvXNOnTqFNm3awMbGBlWrVsXs2bMVtZMFBBERkZm4d+8eWrVqBSsrK2zfvh1//vknvvrqK5QtW1Z3zuzZs7Fw4UIsW7YMhw8fhp2dHQICAvDkyRPdOUFBQTh79iyio6Oxbds27N27FyNGjNAdT0tLg7+/Pzw8PBAfH485c+Zg2rRpWL58eaHbyiEMIiIiAyX123V4eDiqVq2KVatW6fbVqFFD9/+yLGP+/PmYPHkyevToAQD47rvv4OLigq1bt6Jfv344d+4coqKicPToUTRr1gwAsGjRInTp0gVffvkl3N3dsWbNGmRkZGDlypWwtrZGgwYNcPLkScydO1ev0CgIeyCIiIgMiFyJUqPRIC0tTW/TaDR5vm9kZCSaNWuGPn36oFKlSmjatCn+97//6Y4nJSUhOTkZfn5+un1OTk5o2bIl4uLiAABxcXFwdnbWFQ8A4OfnB5VKhcOHD+vOadu2LaytrXXnBAQEICEhAffuFW4yMwsIIiKiYhQWFgYnJye9LSwsLM9zL168iKVLl6J27drYsWMHRo4ciffffx8REREAgOTkp3cYuri46L3OxcVFdyw5ORmVKlXSO25paYly5crpnZNXxrPvYQyHMIiIiAyIvAcjNDQUwcHBevvU6rxvC9ZqtWjWrBlmzpwJAGjatCnOnDmDZcuWYeDAgQJbVXTsgSAiIjKgErip1Wo4OjrqbfkVEG5ubqhfv77evnr16uHKlafr6bi6ugIAbt68qXfOzZs3dcdcXV1x69YtveNZWVlISUnROyevjGffwxhFBcTYsWOxb98+JS/JU57jQRkZRc4lIiJ6kbVq1QoJCQl6+/7++294eHgAeDqh0tXVFbt379YdT0tLw+HDh+Ht/XRhN29vb9y/fx/x8fG6c/bs2QOtVouWLVvqztm7dy8yM/9dMCw6Ohp16tTRu+OjIIoKiCVLlqBdu3Z46aWXEB4eXuhxEkN5jQfNXr7GpCwiIiLRRPZAKDFhwgQcOnQIM2fOxPnz57F27VosX74co0ePBgBIkoTx48fj888/R2RkJE6fPo0BAwbA3d0dgYGBAJ72WHTq1AnDhw/HkSNHcODAAYwZMwb9+vWDu7s7AKB///6wtrbG0KFDcfbsWWzYsAELFizINdRi7BopsnPnTt2tINWqVUOPHj2wbds2aLXaQmeEhoYiNTVVb5s4IkhpU4iIiIqFyLswlGjevDm2bNmCdevWoWHDhpgxYwbmz5+PoKB/f0ZOnDgRY8eOxYgRI9C8eXM8fPgQUVFRsLH5d6n6NWvWoG7duujQoQO6dOmC1q1b663x4OTkhJ07dyIpKQleXl744IMPMGXKlELfwgkAkizLhf54KpVKN7szMzMTW7ZswcqVK7Fr1y64uLhg0KBBGDx4MDw9PQvdgByav35X/JqCiHwWhna/2Gdh4D/2LAz5iuhnYbQXlnWot9hnYbT83ldontk/C0Ow/9KzMEQT/SwMyU7s9cs6tl1onm3X8ULzDC2uKu5ZJWOu/iAsy5yYPInSysoKffv2RVRUFC5evIjhw4djzZo1qFOnjsj2ERERkRkSchdGtWrVMG3aNCQlJSEqKkpEJBERUYkpqTkQLxJF60B4eHjAwsIi3+OSJKFjx45FbhQREVFJKs0/+EVRVEAkJSUVVzuIiIjoBcKVKImIiAxICu+e+C9iAUFERGSAQxjG8RoRERGRYmbTA6E9U/Qlsp8luXgIzRNJmyR2XQTJyVloHsrYCY3LviH4/nQ5W1iWh+t9YVkAcOODrULzKq+rLTQPgtdGkOXCLyBXqLwH4r5WtNlZwrIAQOVUUWhedvIFoXnybbH/rlg28Reap40/KjQPXcXGGeJv18aZTQFBRERkLpSuIPlfxAKCiIjIgMjHeZdW7KUhIiIixdgDQUREZIC/XRvHAoKIiMgA50AYxyKLiIiIFGMPBBERkQFOojSOBQQREZEBFTiGYQyHMIiIiEgxxQXE4sWLMWDAAKxfvx4A8P3336N+/fqoW7cuJk2ahKws46u/aTQapKWl6W2aTLGrxhEREZlKJYvbSitFBcTnn3+OSZMm4dGjR5gwYQLCw8MxYcIEBAUFYeDAgVixYgVmzJhhNCcsLAxOTk5625wte03+EERERCKpBG6llaI5EKtXr8bq1avxxhtv4I8//oCXlxciIiIQFBQEAKhbty4mTpyI6dOnF5gTGhqK4OBgvX3aX+cqbDoRERGVFEUFxPXr19GsWTMAQOPGjaFSqdCkSRPd8VdeeQXXr183mqNWq6FWq/X2PbbifE4iIjIPvAvDOEW9K66urvjzzz8BAImJicjOztb9GQDOnj2LSpUqiW0hERHRc6aSZWFbaaXo1/6goCAMGDAAPXr0wO7duzFx4kR8+OGHuHv3LiRJwhdffIHevXsXV1uJiIiei9I8d0EURQXE9OnTYWtri7i4OAwfPhwff/wxGjdujIkTJ+LRo0fo3r17oSZREhER0YtNUQGhUqkwadIkvX39+vVDv379hDaKiIioJHEOhHGcuUhERGSAK1Eax2EeIiIiUow9EERERAZK8wqSokiybB73mGTeTBAbqNWKy7JSGz9HAe31RKF5UoUqQvOEXjsA2rtXheapKnoIy8o++quwLABQNWwrNs+Jt0VTIWVqxOYJ/ndPTrsjNM+6ZguheYZ+cxE3t6/LzfXCsswJhzCIiIhIMQ5hEBERGeBv18axgCAiIjLAAsI4XiMiIiJSjD0QREREBiSuA2EUCwgiIiID7J43jgUEERGRAfZAGKe4gLhx4waWLl2K/fv348aNG1CpVKhZsyYCAwMxaNAgWFhYFEc7iYiIyIwo6qU5duwY6tWrh99++w2ZmZlITEyEl5cX7Ozs8OGHH6Jt27Z48OCB0RyNRoO0tDS9TaPJMPlDEBERiaSCLGwrrRQVEOPHj8eECRNw7Ngx7Nu3D6tXr8bff/+N9evX4+LFi3j06BEmT55sNCcsLAxOTk56W/jCb0z+EERERCKpJHFbaaVoKesyZcrgzJkzqFmzJgBAq9XCxsYGV69ehYuLC6KjozFo0CBcu3atwByNRgONRn/ZVdX9y1CrrU34CPngUtam41LWJuNS1lRiuJS1UHtd+wjLapu8SViWOVE0B6JSpUq4ceOGroC4efMmsrKy4OjoCACoXbs2UlJSjOao1Wqo1fpfnJmPBRYPRERERcBJlMYpGsIIDAzEe++9h6ioKMTExCAoKAg+Pj6wtbUFACQkJKBy5crF0lAiIqLnRSVwK60U9UB8/vnnuHHjBrp3747s7Gx4e3vjhx9+0B2XJAlhYWHCG0lERETmRVEBYW9vjw0bNuDJkyfIysqCvb293nF/f3+hjSMiIioJksQhDGNMWkjKxsZGdDuIiIjMhooFhFGleXiGiIiIigmXsiYiIjJQipdvEIYFBBERkQEOYRhnPgWEhZXgPLFxIslZYhd8kbKzhObBQvCXheD2Sda24rKq1hWWBQCqMk5C84gKTWXeI9KyLHaBuuLGSZTGmfdXHBEREZkl8+mBICIiMhMcwjCOBQQREZEBibMojeIQBhERESnGHggiIiIDnERpnEkFREZGBrZu3Yq4uDgkJycDAFxdXfHaa6+hR48esLbmkzWJiOjFxTkQxikewjh//jzq1auHgQMH4sSJE9BqtdBqtThx4gQGDBiABg0a4Pz588XRViIiIjITinsgRo4ciUaNGuHEiRNwdHTUO5aWloYBAwZg9OjR2LFjh7BGEhERPU+cRGmc4gLiwIEDOHLkSK7iAQAcHR0xY8YMtGzZUkjjiIiISgLnQBineAjD2dkZly5dyvf4pUuX4OzsXGCGRqNBWlqa3qbRiF2dkYiIiIqP4gJi2LBhGDBgAObNm4dTp07h5s2buHnzJk6dOoV58+Zh0KBBGDFiRIEZYWFhcHJy0tvCFywz+UMQERGJpJJkYVtpJcmyrPjThYeHY8GCBUhOTob0/wNFsizD1dUV48ePx8SJEwt8vUajydXjoHpwDWq1WmlTXkjZV84IzVNV9BCaJ/pZGNrkC0LzLKo1FJYl+u/Cwq220DxY/Te+J0iA7EyxeYKfT6RNvSU0T13rVaF5hs7V7iIsq17ib8KyzIlJPyk++ugjfPTRR0hKStK7jbNGjRqFer1arc5VLGRm3DGlKURERMJxEqVxRVqJskaNGvD29oa3t7eueLh69SqGDBkipHFERERknoQvZZ2SkoKIiAjRsURERM+NpJKFbaWV4iGMyMjIAo9fvHjR5MYQERGZA97GaZziAiIwMBCSJKGguZcSB4+IiIhKNcVDGG5ubti8ebNuCWvD7fjx48XRTiIioueGt3Eap7iA8PLyQnx8fL7HjfVOEBERmTtJJW4rrRQPYYSEhCA9PT3f456enoiJiSlSo4iIiMi8KS4g2rRpU+BxOzs7+Pj4mNwgYTIFLo0teDEfVaXCrZdRUiTRJXOW4GXKtdnistJTxWUB0D4Sm6dyqiQ0j0oxwQs//deZwyTKWbNmITQ0FOPGjcP8+fMBAE+ePMEHH3yA9evXQ6PRICAgAF9//TVcXFx0r7ty5QpGjhyJmJgY2NvbY+DAgQgLC4Ol5b8/8mNjYxEcHIyzZ8+iatWqmDx5MgYNGqSofaW4c4WIiMg0JX0b59GjR/HNN9/g5Zdf1ts/YcIE/PLLL9i0aRN+//13XL9+HW+88YbueHZ2Nrp27YqMjAwcPHgQERERWL16NaZMmaI7JykpCV27doWvry9OnjyJ8ePHY9iwYYqfos0CgoiIyIw8fPgQQUFB+N///oeyZcvq9qempuLbb7/F3Llz0b59e3h5eWHVqlU4ePAgDh06BADYuXMn/vzzT/zwww9o0qQJOnfujBkzZmDJkiXIyMgAACxbtgw1atTAV199hXr16mHMmDHo3bs35s2bp6idLCCIiIgMSJK4TekTqEePHo2uXbvCz89Pb398fDwyMzP19tetWxfVqlVDXFwcACAuLg6NGjXSG9IICAhAWloazp49qzvHMDsgIECXUVgsIIiIiAyIHMLI6wnUYWFheb7v+vXrcfz48TyPJycnw9raGs7Oznr7XVxcdM+lSk5O1iseco7nHCvonLS0NDx+/LjQ10jsYxeJiIhKAZFLUIeGhiI4OFhvX15Pn7569SrGjRuH6Oho2NjYCHv/4iK8B+LmzZv47LPPRMcSERG9kNRqNRwdHfW2vAqI+Ph43Lp1C6+88gosLS1haWmJ33//HQsXLoSlpSVcXFyQkZGB+/fv673u5s2bcHV1BfD0ydg3b97MdTznWEHnODo6wtbWttCfS3gBkZycjOnTp4uOJSIiem5EzoEorA4dOuD06dM4efKkbmvWrBmCgoJ0/29lZYXdu3frXpOQkIArV67A29sbAODt7Y3Tp0/j1q1bunOio6Ph6OiI+vXr6855NiPnnJyMwlI8hHHq1KkCjyckJCiNJCIiMisl8RRNBwcHNGzYUG+fnZ0dypcvr9s/dOhQBAcHo1y5cnB0dMTYsWPh7e2NV199FQDg7++P+vXr45133sHs2bORnJyMyZMnY/To0bpej/feew+LFy/GxIkTMWTIEOzZswcbN27Er7/+qqi9iguIJk2a5Ltcdc5+Yw/T0mg0uWagqjSaPLt0iIiI6Kl58+ZBpVKhV69eegtJ5bCwsMC2bdswcuRIeHt7w87ODgMHDtSbWlCjRg38+uuvmDBhAhYsWIAqVapgxYoVCAgIUNQWSVb44IoKFSpg9uzZ6NChQ57Hz549i+7duyM7O//VAqdNm5ZrmGNyyPuYMnGckqYUzIxXopSf5L8UuDmQLMTOrc2+ekZonkX1JsKyshMOCcsCAMm9ttA8rkRJpYU29ZbxkxRQ13pVaJ6ha97thWVVjtsjLMucKP5J4eXlhevXr8PDwyPP4/fv3zf6MK28ZqSqHlxT2hQiIqJiYQ5LWZs7xQXEe++9V+DDtKpVq4ZVq1YVmKFWq3MNV2Rm3FHaFCIiIiohiguInj17Fni8bNmyGDhwoMkNIiIiKmml+THcogi/RFevXsWQIUNExxIRET03Jf0wrReB8AIiJSUFERERomOJiIjIjCgewoiMjCzw+MWLF01uDBERkTngEIZxiguIwMDAfNeByGFsHQgiIiKzxrswjFJcY7m5uWHz5s3QarV5bsePHy+OdhIRET03kkrcVlop/mheXl6Ij4/P97ix3gkiIiJ68SkewggJCSlwHQhPT0/ExMQUqVFEREQlqTT3HIiieCnr4vJ459fGT1JAVesVYVnaI9uFZQHArgl/C83z+7qR0LxTY44JzatU8YHQPNepvsKypgWfFpYFAFOGWQjNs+z7ntA8SV1GaB5UYj+vyGXeVWVdhWUBALT5L89vUtyDu0LzsjcvF5pn+dYEoXkXfD8Smlcv8TeheYbuBPgIy6qw43dhWeaENRYREREpJvapSURERKUBf702igUEERGRAc6BMI6XiIiIiBQzuYD4559/8PDhw1z7MzMzsXfv3iI1ioiIqESpBG6llOKPduPGDbRo0QIeHh5wdnbGgAED9AqJlJQU+PqKmyVPRET0vHEhKeMUf7SPP/4YKpUKhw8fRlRUFP7880/4+vri3r17unPM5M5QIiIiKiaKC4hdu3Zh4cKFaNasGfz8/HDgwAG4ubmhffv2SElJAcBnYRAR0QuOQxhGKf5oqampKFu2rO7ParUamzdvRvXq1eHr64tbt24ZzdBoNEhLS9PbNBmZSptCRERULCSVJGwrrRQXEDVr1sSpU6f09llaWmLTpk2oWbMmunXrZjQjLCwMTk5OetucDTuVNoWIiKh4sAfCKMUfrXPnzli+PPeSqTlFRJMmTYzOgQgNDUVqaqreFvKmv9KmEBERUQlRvJDUF198gUePHuUdZmmJn376CdeuXSswQ61WQ61W6+17bG2ltClERETFojQPPYiiuAfC0tISjo6O+R6/ceMGpk+fXqRGERERlSgOYRgl/KOlpKQgIiJCdCwRERGZEcVDGJGRkQUev3jxosmNISIiMgscwjBKcQERGBgISZIKnCjJdSCIiOhFxjkQxikewnBzc8PmzZuh1Wrz3I4fP14c7SQiIiIzoriA8PLyQnx8fL7HjfVOEBERmT1OojRK8RBGSEgI0tPT8z3u6emJmJiYIjWKiIioRHEIwyjFBUSbNm0KPG5nZwcfHx/lLbl3W/lrCqA9tkNYVta+I8KyAKCSylZoXvbBw0LzLsn2QvNcLdKE5ol8vF0PTYawLACQnFyE5skPU8TmPbovNE9S2wnNk+8ni8uyUPzPW4Ek+7LGT1IiUyM0Tqr/stg8G7F/t7cfiM2rJzSNTCH2O4yIiKgU4CRK41hAEBERGWIBYRQLCCIiIkMsIIwqxfNDiYiIqLiwB4KIiMgAF0Q0zqQC4u7duzh16hQaN26McuXK4c6dO/j222+h0WjQp08f1KvH+bFERPQC4xCGUYoLiCNHjsDf3x9paWlwdnZGdHQ0+vTpA0tLS2i1WsyaNQv79+/HK6+8UhztJSIiIjOgeA7EJ598gj59+iA1NRWTJk1CYGAgOnTogL///hvnz59Hv379MGPGjOJoKxER0fOhksRtpZTiAiI+Ph7BwcFwcHDAuHHjcP36dQwfPlx3fMyYMTh69GiBGRqNBmlpaXqbJjNLeeuJiIiKg0olbiulFH+yjIwM2No+XUnRysoKZcqUQYUKFXTHK1SogLt37xaYERYWBicnJ71tztZ9SptCREREJURxAVG1alVcvHhR9+f169fDzc1N9+cbN27oFRR5CQ0NRWpqqt4WEljwEtlERETPDYcwjFI8ibJfv364deuW7s9du3bVOx4ZGYkWLVoUmKFWq6FWq/X2PbbiHaVERGQeuJS1cYp/ak+dOrXA45988gksLCxMbhARERGZP+GzO+7evYuRI0eKjiUiInp+OIRhlPACIiUlBREREaJjiYiInh9JJW4rpRQPYURGRhZ4/NkJlkRERC+kUtxzIIriAiIwMBCSJEGW5XzP4RriREREpZvivhU3Nzds3rwZWq02z+348ePF0U4iIqLnh3MgjFJcQHh5eSE+Pj7f48Z6J4iIiMydpFIJ20orxUMYISEhSE9Pz/e4p6cnYmJiitQoIiIiMm+KC4g2bQpeMdLOzg4+Pj6KGyJ51FX8moLICSfEhVmKXdeidsuCl/pWSirrITSvuvRQaN7jh9ZC8+TLScKygjLOC8sCgDOptkLzVOXcheZJajuhebBSGz9HAa2FlbAsydZBWFZxkOycxQaWsRebl50pNK6s7ROhecWuFA89iMLlH4mIiAyV4tsvReEVIiIiIsXYA0FERGSIQxhGsYAgIiIyVIrvnhBF2BWqWbMmEhMTRcURERGRGVPcA7Fw4cI891+5cgWrVq2Cq6srAOD9998vWsuIiIhKCocwjFJcQIwfPx6VK1eGpaX+S7VaLb777jtYWVlBkiQWEERE9OLiXRhGKS4gRowYgcOHD2Pt2rWoV6+ebr+VlRV27tyJ+vXrC20gERHRc8ceCKMUl1jLli3DlClTEBAQgMWLF5v0phqNBmlpaXqbJkPsoiVERERUfEzqo+nZsyfi4uKwZcsWdO7cGcnJyYpeHxYWBicnJ71tzndbTWkKERGRcHwWhnEmf7LKlStj165daNu2LZo2baroAVqhoaFITU3V20IGBJraFCIiIrH4NE6jirQOhCRJCA0Nhb+/P/bv3w83N7dCvU6tVkOt1l9D/4m1uDXwiYiIqHgJ6Vvx8vLCuHHjULZsWVy9ehVDhgwREUtERFQyJJW4rZQS/slSUlIQEREhOpaIiOj54RCGUYqHMCIjIws8fvHiRZMbQ0RERC8GxQVEYGAgJEkqcNKkJJXeiouIiP4DSvHdE6IovkJubm7YvHkztFptntvx48eLo51ERETPjySJ20opxQWEl5cX4uPj8z1urHeCiIiI8hYWFobmzZvDwcEBlSpVQmBgIBISEvTOefLkCUaPHo3y5cvD3t4evXr1ws2bN/XOuXLlCrp27YoyZcqgUqVKCAkJQVZWlt45sbGxeOWVV6BWq+Hp6YnVq1craqviAiIkJASvvfZavsc9PT0RExOjNJaIiMh8qFTiNgV+//13jB49GocOHUJ0dDQyMzPh7++P9PR03TkTJkzAL7/8gk2bNuH333/H9evX8cYbb+iOZ2dno2vXrsjIyMDBgwcRERGB1atXY8qUKbpzkpKS0LVrV/j6+uLkyZMYP348hg0bhh07dhS6rZJsJt0FD97rJDTPsnegsKyMNT8KywIAyVLs2Fp2SobQvMc3xHa5xVyqLDSve+ebxk8qpHLf/yksCwBuBngKzSvTv53QPLhWFRqnKi82T/v3UWFZUu1XhGUBxfBZL58Sm3dwt9A8q8GTheYdf/lDoXktr28Wmmfo0ZIxwrLKjDbtsQ8AcPv2bVSqVAm///472rZti9TUVFSsWBFr165F7969AQB//fUX6tWrh7i4OLz66qvYvn07unXrhuvXr8PFxQXA08dQfPTRR7h9+zasra3x0Ucf4ddff8WZM2d079WvXz/cv38fUVFRhWobZ4kQEREZEngbZ57Pf9JoCtWM1NRUAEC5cuUAAPHx8cjMzISfn5/unLp166JatWqIi4sDAMTFxaFRo0a64gEAAgICkJaWhrNnz+rOeTYj55ycjEJdokKfSURERIrl9fynsLAwo6/TarUYP348WrVqhYYNGwIAkpOTYW1tDWdnZ71zXVxcdM+lSk5O1iseco7nHCvonLS0NDx+/LhQn6tIS1kTERGVSgJXkAwNDUVwcLDePsPHOeRl9OjROHPmDPbv3y+sLSKxgCAiIjIkcB2IvJ7/ZMyYMWOwbds27N27F1WqVNHtd3V1RUZGBu7fv6/XC3Hz5k24urrqzjly5IheXs5dGs+eY3jnxs2bN+Ho6AhbW9tCtZFDGERERGZClmWMGTMGW7ZswZ49e1CjRg29415eXrCyssLu3f9Omk1ISMCVK1fg7e0NAPD29sbp06dx69Yt3TnR0dFwdHRE/fr1dec8m5FzTk5GYRS5B0KWZcTGxuL8+fNwc3NDQEAArKz4ZE0iInpxSSX0DIvRo0dj7dq1+Pnnn+Hg4KCbs+Dk5ARbW1s4OTlh6NChCA4ORrly5eDo6IixY8fC29sbr776KgDA398f9evXxzvvvIPZs2cjOTkZkydPxujRo3U9Ie+99x4WL16MiRMnYsiQIdizZw82btyIX3/9tdBtVVxAdOnSBevWrYOTkxNSUlLQpUsXHDlyBBUqVMDdu3fx0ksvYe/evahYsaLSaCIiIvNQQk/RXLp0KQCgXbt2evtXrVqFQYMGAQDmzZsHlUqFXr16QaPRICAgAF9//bXuXAsLC2zbtg0jR46Et7c37OzsMHDgQHz22We6c2rUqIFff/0VEyZMwIIFC1ClShWsWLECAQEBhW6r4gIiKipKd/vJ5MmT8eDBA1y4cAE1atTAP//8g8DAQEyZMkV3EYiIiKhwCrM0k42NDZYsWYIlS5bke46Hhwd+++23AnPatWuHEydOKG5jjiINYezZswezZ8/WjdFUqVIF4eHhGD58eIGv02g0ue6BzcjWQm3BKRlERGQG+DAto0y6QjlP27x37x5q1aqld8zT0xPXr18v8PV53RP71Qk+BpyIiMxECS1l/SIx6ZMNGjQIb7zxBjIzM5GUlKR3LDk5OdcCF4ZCQ0ORmpqqt33QtKYpTSEiIqISoHgIY+DAgbr/79GjBx49eqR3/KeffkKTJk0KzMjrntgHHL4gIiJzUYofwy2K4gJi1apVBR6fOnUqLCwsTG4QERFRiSvFQw+iCL9CKSkpGDVqlOhYIiKi50dSidtKqWIpICIiIkTHEhERkRlRPIQRGRlZ4PGLF3k3BRERveA4hGGU4gIiMDAQkiQVuNiFxMknRET0ImMBYZTiK+Tm5obNmzdDq9XmuR0/frw42klERERmRHEB4eXlhfj4+HyPG+udICIiMnuSJG4rpRQPYYSEhCA9PT3f456enoiJiSlSo4iIiEoUhzCMUlxAtGnTpsDjdnZ28PHxUd6Qzu0Vv6ZA2VnCom4fFruuReWPmgrNs7IU+/h068TzQvMarEoTmmf1Zk9hWTG/OgnLAoAyw1oIzVPVe01onmRtIzbPvpzQPAhsn8qpkrAsAIBFkR4dlIuqWkOhebB1EBonp90RmtdoIp/QXNqI/Y4gIiIqDdgDYRQLCCIiIkOleAEoUXiFiIiISDH2QBARERniEIZRiq/QP//8gzt3/p1cs2/fPgQFBaFNmzZ4++23ERcXJ7SBREREzx2fhWGU4k/Wq1cvHDp0CADw888/o127dnj48CFatWqFR48ewcfHB9u2bRPeUCIioudGpRK3lVKKhzDOnj2LBg0aAADCwsIwc+ZMfPTRR7rjixcvxpQpU9CtWzdxrSQiIiKzorg0srS0xIMHDwAASUlJ6Ny5s97xzp07IyEhQUzriIiISgKHMIxS/Ml8fHywbt06AEDTpk0RGxurdzwmJgaVK1cuMEOj0SAtLU1v02SKW/iJiIioSDiEYZTiIYxZs2ahTZs2uH79Olq3bo1PPvkER48eRb169ZCQkIANGzZg2bJlBWaEhYVh+vTpevsmvemHyW91VNocIiIiKgGKC4h69erh8OHDmDx5MmbPno309HSsWbMGlpaWaN68OdavX4/AwMACM0JDQxEcHKy3T7tzsdKmEBERFY9S3HMgiknrQNSqVQvr1q2DLMu4desWtFotKlSoACurwj2TQa1WQ61W6+17bMUlKYiIyEyU4rkLohTpCkmSBBcXF7i5uemKh6tXr2LIkCFCGkdERETmSXiJlZKSgoiICNGxREREzw8nURqleNwgMjKywOMXL140uTFERERmgUMYRikuIAIDAyFJEmRZzvccSZKK1CgiIiIyb4pLLDc3N2zevBlarTbP7fjx48XRTiIioueHQxhGKf5kXl5eiI+Pz/e4sd4JIiIis8eVKI1SPIQREhKC9PT0fI97enoiJiamSI0iIiIqUaW450AUSTaT7oLMG+fEBgr8y5czngjLAgBtyjWheZJkITQPFmLX5JAf3DF+kgKSs6uwrKw1S4RlAYBF9z5C81RV6gvNkyytheZB1oqNyxa3pL1kbSssq1hos8XmqQT/OyBY1vEooXm2/qOE5hl6vHe1sCzbtoOEZZkTrt5ERERkQPgvZqUQCwgiIiJDHMIwileIiIiIFGMPBBERkSH2QBjFAoKIiMhQKb79UhTFV+irr77C5cuXi6MtRERE9IJQXECEhISgVq1a6NixIzZs2ICMjIziaBcREVHJ4UqURpn0yVasWAE7Ozu88847cHd3x/jx43HmzBnRbSMiIioZXInSKJM+WZcuXbB161b8888/mDhxInbs2IHGjRujRYsW+N///ocHDx4U+HqNRoO0tDS9TaNhTwYREdGLokilUaVKlTBx4kScO3cOsbGxqF+/PiZMmAA3N7cCXxcWFgYnJye9LXzR8qI0hYiISByVhbitlFJ8F0Z+j+pu06YN2rRpg4ULF2LDhg0FZoSGhiI4OFhvnyolSWlTiIiIikcpHnoQRXEBYezRGY6Ojhg+fHiB56jVaqjVar19memC1+gnIiIyVSme/CiK4gJCqxX78BwiIiJ68Qgvsa5evYohQ4aIjiUiInp+eBeGUcI/WUpKCiIiIkTHEhERPTeSZCFsK60UD2FERkYWePzixYsmN4aIiIheDIoLiMDAQEiSVOBkyvzu1CAiInohcBKlUYqvkJubGzZv3gytVpvndvz48eJoJxER0fPDORBGKf5kXl5eiI+Pz/e4sd4JIiIievEpHsIICQlBenp6vsc9PT0RExNTpEYRERGVKA5hGCXJZtJdkHkrUWxgdpa4LCu18XOU0GaLzZPFrs0hi7x2AOSH94TmqZxdhGVpb18RlgUAqnLuQvOEf+2Zu+xMcVkWVuKyqOgyNULjrNzqCc0zpPnrd2FZ6ro+wrLMCUssIiIiUkzxEAYREVGpV4ofgiUKCwgiIiJDpfjuCVFYQBARERniJEqjeIWIiIhIMZMKiG3btmHKlCk4cOAAAGDPnj3o0qULOnXqhOXLlwttIBER0fMmSSphW2ml+JN988036NmzJ3777Td06dIFP/zwAwIDA1G5cmVUr14d48ePx4IFC4qjrURERM+HSiVuK6UUz4FYuHAhvv76awwfPhwxMTHo0qULvvrqK4waNQoA8Oqrr2L27NkYN26c8MYSERGReVBcGiUlJSEgIAAA4Ovri+zsbLRt21Z3vF27drh8+XKBGRqNBmlpaXqbRpOhtClERETFg8/CMErxJytfvryuQLh+/TqysrJw5cq/q/ldvnwZ5cqVKzAjLCwMTk5Oelv4wmVKm0JERFQ8VBbitlJK8RBGjx49MHToUAwcOBCRkZEYMGAAPvjgA6hUKkiShJCQEPj7+xeYERoaiuDgYL19qtSrSptCREREJURxAREeHo6MjAysX78er732GhYtWoSFCxeiR48eyMzMhI+PD8LCwgrMUKvVUKv11/jPfGKttClERETFoxQPPYgi7GFaT548QWZmJhwcHEx6PR+mVQR8mJbJ+DAtM8OHaZVeL9jDtDKunxWWZe3eQFiWORFWYtnY2MDBwQFXr17FkCFDRMUSERGRGRLeR5OSkoKIiAjRsURERM8NF5IyTvEciMjIyAKPX7x40eTGEBERmYUSXABqyZIlmDNnDpKTk9G4cWMsWrQILVq0KLH25EdxAREYGAhJklDQ1AlJkorUKCIiohJVQj0HGzZsQHBwMJYtW4aWLVti/vz5CAgIQEJCAipVqlQibcqP4ivk5uaGzZs3Q6vV5rkdP368ONpJRERU6s2dOxfDhw/H4MGDUb9+fSxbtgxlypTBypUrS7ppuSguILy8vBAfH5/vcWO9E0RERGZP4EJSea++nPuulIyMDMTHx8PPz+/fZqhU8PPzQ1xc3PP89IUjK7R37155+/bt+R5/+PChHBsbqzS2UJ48eSJPnTpVfvLkiVllMa9055lz25hnPlnMM788czF16lQZgN42derUXOddu3ZNBiAfPHhQb39ISIjcokWL59TawlNcQJSk1NRUGYCcmppqVlnMK9155tw25plPFvPML89cPHnyRE5NTdXb8iqSXrQCQvEkSiIiIiq8vFZfzkuFChVgYWGBmzdv6u2/efMmXF1di6t5Jiu9N6gSERG9QKytreHl5YXdu3fr9mm1WuzevRve3t4l2LK8sQeCiIjITAQHB2PgwIFo1qwZWrRogfnz5yM9PR2DBw8u6abl8kIVEGq1GlOnTi1UV9DzzGJe6c4z57Yxz3yymGd+eS+iN998E7dv38aUKVOQnJyMJk2aICoqCi4u4p4BJIqwh2kRERHRfwfnQBAREZFiLCCIiIhIMRYQREREpBgLCCIiIlKMBQQREREpZvYFxKBBgxAYGJhrf2xsLCRJwv3794uc9+OPP8LGxgZfffWVSe2TJAmSJMHKygo1atTAxIkT8eTJkyJnubi4oGPHjli5ciW0Wq3iPMPMZ7dOnTqZnDVr1iy9/Vu3bjX5Ee5Xr17FkCFD4O7uDmtra3h4eGDcuHG4e/euSXm3b9/GyJEjUa1aNajVari6uiIgIAAHDhxQlJPXNXt2mzZtmkntA4C4uDhYWFiga9euJmc8K7/vkZLK6d69e75fX/v27YMkSTh16pSidkmShPfeey/XsdGjR0OSJAwaNEhxO/P73jh//nyhXr9s2TI4ODggKytLt+/hw4ewsrJCu3bt9M7N+ffqwoULRnNlWYafnx8CAgJyHfv666/h7OyMf/75p1BtzJGcnIxx48bB09MTNjY2cHFxQatWrbB06VI8evRIUdazmWPHjkXNmjWhVqtRtWpVdO/eXW8RpMJ69u/C2toanp6e+Oyzz/SuLZkfsy8gituKFSsQFBSEpUuX4oMPPjApo1OnTrhx4wYuXryIefPm4ZtvvsHUqVOLlHXp0iVs374dvr6+GDduHLp162byN1NO5rPbunXrTMqysbFBeHg47t27Z9Lrn3Xx4kU0a9YMiYmJWLduHc6fP49ly5bpVl1LSUlRnNmrVy+cOHECERER+PvvvxEZGYl27dopLkievVbz58+Ho6Oj3r4PP/xQcdtyfPvttxg7diz27t2L69evm5xjroYOHYro6Og8f8itWrUKzZo1w8svv6wos2rVqli/fj0eP36s2/fkyROsXbsW1apVM7mteX1v1KhRo1Cv9fX1xcOHD3Hs2DHdvn379sHV1RWHDx/W+yUiJiYG1apVQ61atYzmSpKEVatW4fDhw/jmm290+5OSkjBx4kQsWrQIVapUKfRnvHjxIpo2bYqdO3di5syZOHHiBOLi4jBx4kRs27YNu3btKnRWjkuXLsHLywt79uzBnDlzcPr0aURFRcHX1xejR49WnAf8+3eRmJiIDz74ANOmTcOcOXNMyqLn44VaSEq02bNnY+rUqVi/fj169uxpck7Ob7rA03/o/Pz8EB0djfDw8CJlVa5cGa+88gpeffVVdOjQAatXr8awYcOKlFlUfn5+OH/+PMLCwjB79uwiZY0ePRrW1tbYuXMnbG1tAQDVqlVD06ZNUatWLXzyySdYunRpofPu37+Pffv2ITY2Fj4+PgAADw8PtGjRQnHbnr1eTk5OkCRJyDV8+PAhNmzYgGPHjiE5ORmrV6/GpEmTipxrTrp164aKFSti9erVmDx5sm7/w4cPsWnTJpN+KLzyyiu4cOECNm/ejKCgIADA5s2bUa1atUL/wM9LUb436tSpAzc3N8TGxuLVV18F8LSnoUePHtizZw8OHTqk64mIjY2Fr69vobOrVq2KBQsWYMyYMfD390f16tUxdOhQ+Pv745133lHUzlGjRsHS0hLHjh2DnZ2dbn/NmjXRo0cPmLIU0KhRoyBJEo4cOaKX2aBBAwwZMkRxHqD/dzFy5Ehs2bIFkZGRCA0NNSmPit9/tgfio48+wowZM7Bt27YiFQ+Gzpw5g4MHD8La2lpYZvv27dG4cWNs3rxZWKapLCwsMHPmTCxatEhxN+qzUlJSsGPHDowaNUpXPORwdXVFUFAQNmzYoOgfN3t7e9jb22Pr1q3QaDQmt604bdy4EXXr1kWdOnXw9ttvY+XKlSb9A27OLC0tMWDAAKxevVrvs23atAnZ2dl46623TModMmQIVq1apfvzypUrS3x5X19fX8TExOj+HBMTg3bt2sHHx0e3//Hjxzh8+LCiAgIABg4ciA4dOmDIkCFYvHgxzpw5o9cjURh3797Fzp07MXr0aL0f9M9SOvyYkpKCqKiofDOdnZ0V5eXH1tYWGRkZQrKoeLwQBcS2bdt0Pxxyts6dO5uct337dsyePRs///wzOnToIKx9NjY2aNSoEW7duoWQkJAi5z6rbt26uHTpUpHa9+w2c+ZMk9vSs2dPNGnSxORhGgBITEyELMuoV69ensfr1auHe/fu4fbt24XOtLS0xOrVqxEREQFnZ2e0atUKkyZNUjTeXty+/fZbvP322wCedtmmpqbi999/L+FWiTdkyBBcuHBB77OtWrUKvXr1gpOTk0mZb7/9Nvbv34/Lly/j8uXLOHDggO5amsrwe6NPnz6KXu/r64sDBw4gKysLDx48wIkTJ+Dj44O2bdsiNjYWwNM5LxqNRnEBAQDLly/HmTNnMH78eCxfvhwVK1ZU9Prz589DlmXUqVNHb3+FChV0n/mjjz4yKbNu3bqKXldYsixj165d2LFjB9q3b18s70FivBBDGL6+vrm6sg8fPmzyPx4vv/wy7ty5g6lTp6JFixawt7cX0r709HTMmzcPlpaW6NWrV5EyDcmybPJExbyuX7ly5YrUnvDwcLRv375IcwEAGP3tW2lPTq9evdC1a1fs27cPhw4d0hWLK1asMGminUgJCQk4cuQItmzZAuBpwfPmm2/i22+/zTXp7kVXt25dvPbaa1i5ciXatWuH8+fPY9++ffjss89MzqxYsSK6du2q69no2rUrKlSoUKR2Gn5v5Pdben7atWuH9PR0HD16FPfu3cNLL72EihUrwsfHB4MHD8aTJ08QGxuLmjVrmjRXo1KlSnj33XexdetWIRNccxw5cgRarRZBQUGKe+uKq8csp5jLzMyEVqtF//79izRZmYrfC1FA2NnZwdPTU29fUbrPK1eujB9//BG+vr7o1KkTtm/fDgcHByHtW7lyJRo3boxvv/0WQ4cONTnT0Llz50we683r+hVV27ZtERAQgNDQUJN+MHt6ekKSJJw7dy7PIaRz586hYsWKJnWH2tjYoGPHjujYsSM+/fRTDBs2DFOnTi3xAuLbb79FVlYW3N3ddftkWYZarcbixYtN/s3cXA0dOhRjx47FkiVLsGrVKtSqVUs3N8VUQ4YMwZgxYwAAS5YsKXIbi/q94enpiSpVqiAmJgb37t3TfT53d3dUrVoVBw8eRExMTJF+k7a0tISlpWn/VOd8nyUkJOjtr1mzJgDkGj4sjNq1a0OSJPz1118mtSk/OcWctbU13N3dTf7M9Py8EEMYxcHDwwO///47kpOT0alTJzx48EBIrkqlwqRJkzB58mS9GeNFsWfPHpw+fVp4r0ZRzZo1C7/88gvi4uIUv7Z8+fLo2LEjvv7661zXKTk5GWvWrBH2A79+/fpIT08XkmWqrKwsfPfdd/jqq69w8uRJ3fbHH3/A3d3d5LtizFnfvn2hUqmwdu1afPfddxgyZIjJvWg5OnXqhIyMDGRmZuZ5m2NJ8PX1RWxsLGJjY/V6ktq2bYvt27fjyJEjJg1fiJDzfbZ48WJh3wPlypVDQEAAlixZkmem0lvrc+QUc9WqVWPx8IL4zxYQwNOZzrGxsbh16xYCAgKQlpYmJLdPnz6wsLAw6TckjUaD5ORkXLt2DcePH8fMmTPRo0cPdOvWDQMGDDCpPTmZz2537twxKetZjRo1QlBQEBYuXGjS6xcvXgyNRoOAgADs3bsXV69eRVRUFDp27IiXXnoJU6ZMUZR39+5dtG/fHj/88ANOnTqFpKQkbNq0CbNnz0aPHj1MaqMo27Ztw7179zB06FA0bNhQb+vVqxe+/fbbEm1fjtTUVL0C5+TJk7h69apJWfb29njzzTcRGhqKGzduCCkILSwscO7cOfz555+wsLAocp4Ivr6+2L9/P06ePKnXw+Lj44NvvvkGGRkZJVZAAE/XjsjKykKzZs2wYcMGnDt3DgkJCfjhhx/w119/mXQdlyxZguzsbLRo0QI//fQTEhMTce7cOSxcuBDe3t7F8CnIHP2nCwgAqFKlCmJjY3Hnzh1hRYSlpSXGjBmD2bNnK676o6Ki4ObmhurVq6NTp06IiYnBwoUL8fPPP5v8D2ZO5rNb69atTcoy9Nlnn5m8yFXt2rVx9OhR1KxZE3379oWHhwc6d+6Ml156CQcOHFA8N8Xe3h4tW7bEvHnz0LZtWzRs2BCffvophg8fjsWLF5vURlG+/fZb+Pn55TlM0atXLxw7dswsJnvGxsaiadOmetv06dNNzhs6dCju3buHgIAAvaGbonB0dISjo6OQLBF8fX3x+PFjeHp6wsXFRbffx8cHDx480N3uWVJq1aqFEydOwM/PD6GhoWjcuDGaNWuGRYsW4cMPP8SMGTMUZ9asWRPHjx+Hr68vPvjgAzRs2BAdO3bE7t27Fd16TS82SS5t95DRC23q1KmYO3cuoqOjdffWExGR+WEBQWZn1apVSE1Nxfvvvw+V6j/fSUZEZJZYQBAREZFi/PWOiIiIFGMBQURERIqxgCAiIiLFWEAQERGRYiwgiIiISDEWEERERKQYCwgiIiJSjAUEERERKcYCgoiIiBT7P4AaFpMQLMigAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# prepare the matrix to plot the heatmap\n",
    "most_sel[\"difference_in_seq\"] = [diff_func(x) for x in most_sel[\"Sequence\"]]\n",
    "diff = most_sel[\"difference_in_seq\"].str.split(\"\", n= -1, expand = True)\n",
    "most_sel= most_sel.merge(diff, left_index=True, right_index=True)\n",
    "\n",
    "# charged->polar->aliphatic->aromatic->other\n",
    "aa_list = ['H', 'K', 'R', 'D', 'E', 'N', 'Q', 'S', 'T', 'A', 'I', 'L', 'V', 'M', 'F', 'W', 'Y','G', 'C', 'P']\n",
    "\n",
    "#aa_list = ['H', 'K', 'R', 'D', 'E', 'A', 'I', 'L', 'V', 'M', 'F', 'W', 'Y', 'N', 'Q', 'S', 'T', 'G', 'C', 'P']\n",
    "mtx = pd.DataFrame(0, index=aa_list, columns=np.arange(1, 19))\n",
    "\n",
    "# fill the matix with the count of each amino acid in each position\n",
    "for i in range(1, 19):\n",
    "    pos_count = most_sel[i].value_counts().to_dict()\n",
    "\n",
    "    for k,v in pos_count.items():\n",
    "        if k != '-':\n",
    "            mtx.loc[k,i] = v\n",
    "\n",
    "\n",
    "# plot the heatmap\n",
    "plt.title(\"Most Selective Peptides\")\n",
    "sns.heatmap(mtx.T, cmap=\"rocket_r\")\n",
    "#plt.savefig('results/changes_heatmap_predictions_most_selective.png' , dpi=1200)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Bacterial Specific (MIC<=8, %hemo<=2, L10SS <1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# prepare the matrix to plot the heatmap\n",
    "bac[\"difference_in_seq\"] = [diff_func(x) for x in bac[\"Sequence\"]]\n",
    "diff = bac[\"difference_in_seq\"].str.split(\"\", n= -1, expand = True)\n",
    "bac = bac.merge(diff, left_index=True, right_index=True)\n",
    "\n",
    "\n",
    "# charged->polar->aliphatic->aromatic->other\n",
    "aa_list = ['H', 'K', 'R', 'D', 'E', 'N', 'Q', 'S', 'T', 'A', 'I', 'L', 'V', 'M', 'F', 'W', 'Y','G', 'C', 'P']\n",
    "\n",
    "#aa_list = ['H', 'K', 'R', 'D', 'E', 'A', 'I', 'L', 'V', 'M', 'F', 'W', 'Y', 'N', 'Q', 'S', 'T', 'G', 'C', 'P']\n",
    "mtx = pd.DataFrame(0, index=aa_list, columns=np.arange(1, 19))\n",
    "\n",
    "# fill the matix with the count of each amino acid in each position\n",
    "for i in range(1, 19):\n",
    "    pos_count = bac[i].value_counts().to_dict()\n",
    "\n",
    "    for k,v in pos_count.items():\n",
    "        if k != '-':\n",
    "            mtx.loc[k,i] = v\n",
    "\n",
    "# plot the heatmap\n",
    "plt.title(\"Bacterial Specific Peptides\")\n",
    "sns.heatmap(mtx.T, cmap=\"rocket_r\")\n",
    "#plt.savefig('results/changes_heatmap_predictions_bac.png', dpi=1200)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Mammalian specific (MIC>8, %hemo>50, L10SS >1). "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhAAAAGzCAYAAAB+YC5UAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/OQEPoAAAACXBIWXMAAA9hAAAPYQGoP6dpAABY3ElEQVR4nO3deVxU1fsH8M8dlgFBwJXFBVFSwV1MJUNESVQ0MVtITQ2zXHMpNNLcKlH6mvu+kmmaZWb4FVfIJVwCyzQ0DXcFFxQUdVjm/v7ox3wdtuEOBxmnz7vXeb3i3jvPPPfK8sw5554rybIsg4iIiEgBVUUnQERERM8eFhBERESkGAsIIiIiUowFBBERESnGAoKIiIgUYwFBREREirGAICIiIsVYQBAREZFiLCCIiIhIMRYQZDbi4+MhSRLi4+N12wYPHox69epVWE6mrFOnTujUqZPetrS0NLz66quoVq0aJEnCvHnziryu5mTatGmQJKlUx0qShGnTppVvQkTPCBYQz7h169ZBkiRIkoRDhw4V2i/LMurUqQNJktCzZ88KyPDf69atWxgzZgwaN24MW1tb1KxZE23btsXEiRPx4MGDik6vSOPGjcOuXbsQERGB9evXo1u3bkLj5xcj+c3Kygr169fHwIEDkZKSIvS9nvTw4UNMmzbNbIsgoopgWdEJkBg2NjbYuHEjXnzxRb3tP//8M65evQq1Wl1BmVWslStXQqvVPvX3TU9PR5s2bZCZmYmwsDA0btwYd+7cwcmTJ7F06VIMHz4c9vb2Tz2vJ+3evbvQtv3796N379748MMPddsaNmyIR48ewdraWth7v//++3j++eeRk5ODpKQkrFixAjt27MAff/wBNzc3Ye+T7+HDh5g+fToAFOp1mTx5Mj766CPh70lk7lhAmIkePXpgy5YtWLBgASwt//fPunHjRvj4+OD27dsVmF3FsbKyqpD3Xb16NS5fvozDhw/jhRde0NuXmZkp9I+xsYrK4ebNm3ByctLbplKpYGNjI/S9/fz88OqrrwIA3n77bTRs2BDvv/8+oqOjERERIfS9DLG0tNT7mSGi0uEQhpl48803cefOHezZs0e3LTs7G9999x369etX5Gv+85//4IUXXkC1atVga2sLHx8ffPfdd4WOkyQJo0aNwpYtW+Dt7Q1bW1v4+vrijz/+AAAsX74cnp6esLGxQadOnXDx4kW913fq1AlNmzbFyZMn4e/vj0qVKsHT01P3Xj///DPatWsHW1tbNGrUCHv37tV7/aVLlzBixAg0atQItra2qFatGl577bVC71OUouZAKD3vbdu2oWnTplCr1WjSpAliY2MNvu/ff/8NCwsLtG/fvtA+BwcHvT/I+dcnMTERL7zwAmxtbeHh4YFly5YVeq1Go8HUqVPh6ekJtVqNOnXqYMKECdBoNIWO/frrr9G2bVtUqlQJVapUQceOHfV6HZ6cA5E/FCbLMhYvXqwbYgCKnlsCAEePHkWPHj1QpUoV2NnZoXnz5pg/f77Ba1OUzp07AwAuXLig27Zz5074+fnBzs4OlStXRnBwME6fPq33usGDB8Pe3h4pKSkICgqCnZ0d3NzcMGPGDOQ/aPjixYuoUaMGAGD69Om6c8ufy1DUHAiNRoNx48ahRo0aqFy5Ml5++WVcvXq1yNyvXbuGsLAwODs7675H1qxZU+i4hQsXokmTJrp/jzZt2mDjxo1GXS8iU8ACwkzUq1cPvr6++Oabb3Tbdu7ciYyMDISGhhb5mvnz56NVq1aYMWMGZs6cCUtLS7z22mvYsWNHoWMPHjyIDz74AIMGDcK0adOQnJyMnj17YvHixViwYAFGjBiB8PBwJCQkICwsrNDr7969i549e6Jdu3aIioqCWq1GaGgoNm/ejNDQUPTo0QOzZs1CVlYWXn31Vdy/f1/32uPHj+OXX35BaGgoFixYgGHDhmHfvn3o1KkTHj58qPhaKTnvQ4cOYcSIEQgNDUVUVBQeP36Mvn374s6dOyW+h7u7O/Ly8rB+/fpS5XT37l306NEDPj4+iIqKQu3atTF8+HC9P0RarRYvv/wy/vOf/6BXr15YuHAhQkJCMHfuXLzxxht68aZPn4633noLVlZWmDFjBqZPn446depg//79Rb5/x44ddbm+9NJLWL9+fYm579mzBx07dsSff/6JMWPGYM6cOQgICEBMTEypzregv//+GwBQrVo1AMD69esRHBwMe3t7zJ49G5988gn+/PNPvPjii4UKx7y8PHTr1g3Ozs6IioqCj48Ppk6diqlTpwIAatSogaVLlwIA+vTpozu3V155pdh83nnnHcybNw9du3bFrFmzYGVlheDg4ELHpaWloX379ti7dy9GjRqF+fPnw9PTE0OGDMG8efN0x61cuRLvv/8+vL29MW/ePEyfPh0tW7bE0aNHjbpeRCZBpmfa2rVrZQDy8ePH5UWLFsmVK1eWHz58KMuyLL/22mtyQECALMuy7O7uLgcHB+u9Nv+4fNnZ2XLTpk3lzp07620HIKvVavnChQu6bcuXL5cByC4uLnJmZqZue0REhAxA71h/f38ZgLxx40bdtjNnzsgAZJVKJR85ckS3fdeuXTIAee3atcXmKcuynJCQIAOQv/rqK922uLg4GYAcFxen2zZo0CDZ3d3d6PO2traWz58/r9v2+++/ywDkhQsXFsrpSampqXKNGjVkAHLjxo3lYcOGyRs3bpTv3btX6Nj86zNnzhzdNo1GI7ds2VKuWbOmnJ2dLcuyLK9fv15WqVTywYMH9V6/bNkyGYB8+PBhWZZl+dy5c7JKpZL79Okj5+Xl6R2r1Wr13tff37/QOY8cOVJvW8HrmpubK3t4eMju7u7y3bt3i41flPxYa9askW/duiVfv35d3rFjh1yvXj1ZkiT5+PHj8v3792UnJyd56NCheq9NTU2VHR0d9bYPGjRIBiCPHj1aL4fg4GDZ2tpavnXrlizLsnzr1i0ZgDx16tRCOU2dOlV+8lfhb7/9JgOQR4wYoXdcv379CsUYMmSI7OrqKt++fVvv2NDQUNnR0VH3vda7d2+5SZMmJV4bomcNeyDMyOuvv45Hjx4hJiYG9+/fR0xMTLHDFwBga2ur+/+7d+8iIyMDfn5+SEpKKnRsly5d9IYC2rVrBwDo27cvKleuXGh7wRn19vb2ej0hjRo1gpOTE7y8vHSvKe71T+aZk5ODO3fuwNPTE05OTkXmaoiS8w4MDESDBg10Xzdv3hwODg4G7xhwdnbG77//jmHDhuHu3btYtmwZ+vXrh5o1a+LTTz/Vda/ns7S0xHvvvaf72traGu+99x5u3ryJxMREAMCWLVvg5eWFxo0b4/bt27qW3/0fFxcHANi2bRu0Wi2mTJkClUr/R7y0tyuW5MSJE7hw4QLGjh1baL5EaeOHhYWhRo0acHNzQ3BwMLKyshAdHY02bdpgz549uHfvHt58802987SwsEC7du105/mkUaNG6eUwatQoZGdnFxoOK43//ve/AP6Z6PmksWPH6n0tyzK+//579OrVC7Is6+UaFBSEjIwM3feUk5MTrl69iuPHjyvOh8hUceaQGalRowYCAwOxceNGPHz4EHl5ebqJakWJiYnBZ599ht9++01vDL2oPwJ169bV+9rR0REAUKdOnSK33717V2977dq1C8V1dHQs1esfPXqEyMhIrF27FteuXdP745uRkVHs+RWnLOcNAFWqVCl0fkVxdXXF0qVLsWTJEpw7dw67du3C7NmzMWXKFLi6uuKdd97RHevm5gY7Ozu91zds2BDAP2P47du3x7lz55CcnKwbzy/o5s2bAP4ZDlCpVPD29jaYozHyhxuaNm1qdIwpU6bAz88PFhYWqF69Ory8vHQTGc+dOwfgf/MiCnJwcND7WqVSoX79+nrbnrx2Sl26dAkqlUqvcAT+KXqfdOvWLdy7dw8rVqzAihUrioyV/28yceJE7N27F23btoWnpye6du2Kfv36oUOHDorzIzIVLCDMTL9+/TB06FCkpqaie/fuhT4h5jt48CBefvlldOzYEUuWLIGrqyusrKywdu3aIid2WVhYFBmnuO0FP2GX5fWjR4/G2rVrMXbsWPj6+sLR0RGSJCE0NFTxLZqizrvg+ZVEkiQ0bNgQDRs2RHBwMJ577jls2LBBr4AoDa1Wi2bNmuHLL78scn/BYsyUNWvWDIGBgUXuy/83Xb9+PVxcXArtN5U7JvLzHDBgAAYNGlTkMc2bNwcAeHl54ezZs4iJiUFsbCy+//57LFmyBFOmTNHdXkr0rDGNn0QSpk+fPnjvvfdw5MgRbN68udjjvv/+e9jY2GDXrl16a0SsXbv2aaSpyHfffYdBgwZhzpw5um2PHz/GvXv3FMeq6POuX78+qlSpghs3buhtv379OrKysvR6If766y8A0A0dNWjQAL///ju6dOlS4lBBgwYNoNVq8eeff6Jly5bCzyH/k/mpU6eKLQJExK9Zs2ap4mu1WqSkpOh6HYDC107J0I27uzu0Wi3+/vtvvV6Hs2fP6h2Xf4dGXl5eqfK0s7PDG2+8gTfeeAPZ2dl45ZVX8PnnnyMiIkL4bbJETwPnQJgZe3t7LF26FNOmTUOvXr2KPc7CwgKSJCEvL0+37eLFi9i2bdtTyFIZCwuLQp/4Fy5cqJe7klhP47yPHj2KrKysQtuPHTuGO3fuFOoOz83NxfLly3VfZ2dnY/ny5ahRowZ8fHwA/DPH5dq1a1i5cmWhuI8ePdK9X0hICFQqFWbMmFGoh0ZJz0lxWrduDQ8PD8ybN69QESciflBQEBwcHDBz5kzk5OQU2n/r1q1C2xYtWqSXw6JFi2BlZYUuXboAACpVqgQApSo6u3fvDgBYsGCB3vYn76oA/vle6tu3L77//nucOnWqxDwL3rVjbW0Nb29vyLJc5DkSPQvYA2GGiutOfVJwcDC+/PJLdOvWDf369cPNmzexePFieHp64uTJk08hy9Lr2bMn1q9fD0dHR3h7eyMhIQF79+7V3fKnxNM67/Xr12PDhg3o06cPfHx8YG1tjeTkZKxZswY2Njb4+OOP9Y53c3PD7NmzcfHiRTRs2BCbN2/Gb7/9hhUrVugWw3rrrbfw7bffYtiwYYiLi0OHDh2Ql5eHM2fO4Ntvv8WuXbvQpk0beHp6YtKkSfj000/h5+eHV155BWq1GsePH4ebmxsiIyPLdG4qlQpLly5Fr1690LJlS7z99ttwdXXFmTNncPr0aezatatM8R0cHLB06VK89dZbaN26NUJDQ1GjRg1cvnwZO3bsQIcOHfQKBhsbG8TGxmLQoEFo164ddu7ciR07duDjjz/WzRextbWFt7c3Nm/ejIYNG6Jq1apo2rRpkfM4WrZsiTfffBNLlixBRkYGXnjhBezbtw/nz58vdOysWbMQFxeHdu3aYejQofD29kZ6ejqSkpKwd+9epKenAwC6du0KFxcXdOjQAc7OzkhOTsaiRYsQHBysNwmZ6FnCAuJfqnPnzli9ejVmzZqFsWPHwsPDQ/cHzNQKiPnz58PCwgIbNmzA48eP0aFDB+zduxdBQUGKYz2t837vvfdQqVIl7Nu3Dz/++CMyMzNRo0YNdO3aFREREWjVqpXe8VWqVEF0dDRGjx6NlStXwtnZGYsWLcLQoUN1x6hUKmzbtg1z587FV199hR9++AGVKlVC/fr1MWbMGL0u/BkzZsDDwwMLFy7EpEmTUKlSJTRv3hxvvfWWkPMLCgpCXFwcpk+fjjlz5kCr1aJBgwZ6+ZZFv3794ObmhlmzZuGLL76ARqNBrVq14Ofnh7ffflvvWAsLC8TGxmL48OEIDw9H5cqVMXXqVEyZMkXvuFWrVmH06NEYN24csrOzMXXq1GIngq5ZswY1atTAhg0bsG3bNnTu3Bk7duwoNM/E2dkZx44dw4wZM7B161YsWbIE1apVQ5MmTTB79mzdce+99x42bNiAL7/8Eg8ePEDt2rXx/vvvY/LkyUKuF1FFkGQRfY5EZLROnTrh9u3bRXaDU8kGDx6M7777zmQfTkZkzjgHgoiIiBRjAUFERESKsYAgIiIixTgHgoiIiBRjDwQREREpxgKCiIiIFGMBQURERIqZzEJS2Vd+r+gUiiXZCl4pTqt8CeaSyA+VP5GyJJLazvBBCsjZj4TGg6roB1wZxULwj4DCh3sZJPraiWYj9ntFsrY1fFAp5V0Wu66G9sAOofEsXnpdaDzR38uqqm5C4yEvV2g4K1cvofEKsrSuJSxWbvY1YbFMickUEERERKai9I9f+/fiEAYREREpxh4IIiKiApQ8Av7figUEERFRASwgDGMBQUREVIDEWRAGcQ4EERERKcYeCCIiogI4hGEYCwgiIqICVCwgDBI+hHHlyhWEhYWVeIxGo0FmZqZe02iyRadCRERE5UR4AZGeno7o6OgSj4mMjISjo6Nei1q8WnQqRERERpEE/meuFA9hbN++vcT9KSkpBmNERERg/Pjxetukm2eVpkJERFQuOIRhmOICIiQkBJIkQZblYo8xNPlErVZDrVbrbcvOsFaaChEREVUQxUMYrq6u2Lp1K7RabZEtKSmpPPIkIiJ6aiRJEtbMleICwsfHB4mJicXuN9Q7QUREZOpUkIQ1c6V4CCM8PBxZWVnF7vf09ERcXFyZkiIiIiLTpriA8PPzK3G/nZ0d/P39jU6IiIioopnz0IMoXEiKiIioAN6FYRgLCCIiogLYA2EYH6ZFREREiplOD4QkuJaRteJCPbovLBYASOpKYuPZVhYaT85+LDSeZGMnNJ6cI3DZc6247xMAgErs97H8MENoPFgKXm8lL1doODnztrhgeTniYgHITb4oNJ7q+VSh8Swa+QqN929nzndPiGI6BQQREZGJ4BCGYRzCICIiIsXYA0FERFSAOT8ESxQWEERERAXwNk7DOIRBREREirEHgoiIqABOojSMBQQREVEBvI3TMA5hEBERkWIV0gOh0Wig0Wj0tkmabKjVghe5ISIiMgKHMAxT3APx6NEjHDp0CH/++WehfY8fP8ZXX31lMEZkZCQcHR31WtTiVUpTISIiKhcqSMKauVJUQPz111/w8vJCx44d0axZM/j7++PGjRu6/RkZGXj77bcNxomIiEBGRoZemzDyHeXZExERlQNJUglr5krRmU2cOBFNmzbFzZs3cfbsWVSuXBkdOnTA5cuXFb2pWq2Gg4ODXuPwBRER0bND0RyIX375BXv37kX16tVRvXp1/PTTTxgxYgT8/PwQFxcHOzuxD00iIiKqCFyJ0jBFPRCPHj2CpeX/ag5JkrB06VL06tUL/v7++Ouvv4QnSERE9LSpJElYM1eKeiAaN26MX3/9FV5eXnrbFy1aBAB4+eWXxWVGREREJktRD0SfPn3wzTffFLlv0aJFePPNNyHLspDEiIiIKook8D9zpaiAiIiIwH//+99i9y9ZsgRarbbMSREREVUkDmEYZr73lxAREVG54bMwiIiICjDnoQdRWEAQEREVYM5DD6KYTAEhP8wQGk9SVxIXLC9XXCwAsK0sNl6e2HknkrWN0HgQvRJbXo64WLLgOTuCz1V75qjQeKhSU2y8Rw+EhpOquQqLdWHgemGxAKDZpVNC46VeXCE0nuMGX6HxtLevCI2H7Mdi49V8Tmw8UsxkCggiIiJTwSEMw1hAEBERFcAhDMNYQBARERXAHgjDeBsnERERKcYeCCIiogJU7IEwiAUEERFRARLnQBjEIQwiIiJSjD0QREREBXAIwzDFBURycjKOHDkCX19fNG7cGGfOnMH8+fOh0WgwYMAAdO7c2WAMjUYDjUajvzE7G2pra6XpEBERCcchDMMUDWHExsaiZcuW+PDDD9GqVSvExsaiY8eOOH/+PC5duoSuXbti//79BuNERkbC0dFRr0Ut/9rokyAiIqKnS1EBMWPGDISHh+POnTtYu3Yt+vXrh6FDh2LPnj3Yt28fwsPDMWvWLINxIiIikJGRodcmvDfA6JMgIiISSRL4n7lSVECcPn0agwcPBgC8/vrruH//Pl599VXd/v79++PkyZMG46jVajg4OOg1Dl8QEZGpUEES1syV4rsw8seFVCoVbGxs4OjoqNtXuXJlZGSIfSgWERERmR5FBUS9evVw7tw53dcJCQmoW7eu7uvLly/D1VXc0/SIiIgqgiRJwpq5UnQXxvDhw5GXl6f7umnTpnr7d+7cWaq7MIiIiEyZOQ89iKKogBg2bFiJ+2fOnFmmZIiIiEwBCwjDuBIlERERKcaVKImIiApg/4NhLCCIiIgKUJnx5EdRTKaAkKxthMaT798RFkvl0kBYLACANs/wMQrI2Y+FxhOdH6zUYuPl5YqNJ5AsC84tT+y/heRUU2g8WNsKDSfnaAwfVEr1Pm8vLBYAnJhoITSe3Qg/ofHykg8LjSfn5QiNZ9GwndB4VPFMpoAgIiIyFea8gqQonERJRERUQEWtRJmXl4dPPvkEHh4esLW1RYMGDfDpp59ClmXdMbIsY8qUKXB1dYWtrS0CAwP11mgCgPT0dPTv3x8ODg5wcnLCkCFD8ODBA71jTp48CT8/P9jY2KBOnTqIiopSeI2IiIjIJMyePRtLly7FokWLkJycjNmzZyMqKgoLFy7UHRMVFYUFCxZg2bJlOHr0KOzs7BAUFITHj/83nN2/f3+cPn0ae/bsQUxMDA4cOIB3331Xtz8zMxNdu3aFu7s7EhMT8cUXX2DatGlYsWJFqXPlEAYREVEBFbWC5C+//ILevXsjODgYwD8rQH/zzTc4duwYgH96H+bNm4fJkyejd+/eAICvvvoKzs7O2LZtG0JDQ5GcnIzY2FgcP34cbdq0AQAsXLgQPXr0wH/+8x+4ublhw4YNyM7Oxpo1a2BtbY0mTZrgt99+w5dffqlXaJSEPRBEREQFiBzC0Gg0yMzM1GsaTdEThl944QXs27cPf/31FwDg999/x6FDh9C9e3cAwIULF5CamorAwEDdaxwdHdGuXTskJCQA+OcxE05OTrriAQACAwOhUqlw9OhR3TEdO3aE9RMPsgwKCsLZs2dx9+7dUl4jIiIiKjeRkZFwdHTUa5GRkUUe+9FHHyE0NBSNGzeGlZUVWrVqhbFjx6J///4AgNTUVACAs7Oz3uucnZ11+1JTU1Gzpv4dV5aWlqhatareMUXFePI9DOEQBhERUQEi78KIiIjA+PHj9bap1UXf3v7tt99iw4YN2Lhxo25YYezYsXBzc8OgQYOE5SSCkAJClmWzfuIYERH9u4jsnler1cUWDAWFh4freiEAoFmzZrh06RIiIyMxaNAguLi4AADS0tL0nn6dlpaGli1bAgBcXFxw8+ZNvbi5ublIT0/Xvd7FxQVpaWl6x+R/nX+MIUKukVqtRnJysohQREREFa6iHuf98OFDqFT6f5otLCyg1WoBAB4eHnBxccG+fft0+zMzM3H06FH4+voCAHx9fXHv3j0kJibqjtm/fz+0Wi3atWunO+bAgQPIyfnfgmF79uxBo0aNUKVKlVLlqqgHomAXTL68vDzMmjUL1apVAwB8+eWXJcbRaDSFJpBImmyo1dbFvIKIiMj89erVC59//jnq1q2LJk2a4MSJE/jyyy8RFhYG4J/CZuzYsfjss8/w3HPPwcPDA5988gnc3NwQEhICAPDy8kK3bt0wdOhQLFu2DDk5ORg1ahRCQ0Ph5uYGAOjXrx+mT5+OIUOGYOLEiTh16hTmz5+PuXPnljpXRQXEvHnz0KJFCzg5Oeltl2UZycnJsLOzK1W1FRkZienTp+ttmzxmKD4ZW7pbR4iIiMpTRT3Oe+HChfjkk08wYsQI3Lx5E25ubnjvvfcwZcoU3TETJkxAVlYW3n33Xdy7dw8vvvgiYmNjYWPzv0dCbNiwAaNGjUKXLl2gUqnQt29fLFiwQLff0dERu3fvxsiRI+Hj44Pq1atjypQppb6FEwAk+cnlrQyYNWsWVqxYgVWrVqFz58667VZWVvj999/h7e1dqjhF9kDcOCW0B0J+ULrbUErD5J+FoXkoNJ7JPwvjcZbYeALJslZoPO2vu4TGkzxbCI2Hh/eFhpNlgd97qZfFxQJwfuJxofE854l9FoZU1dXwQQqY+rMwrN2aCI1XUK+6PYXF+ulyjLBYpkTRHIiPPvoImzdvxvDhw/Hhhx/qjZ0ooVar4eDgoNc4fEFERPTsUDyJ8vnnn0diYiJu3bqFNm3a4NSpU7wDg4iIzIpKYDNXRt3GaW9vj+joaGzatAmBgYHIE/zIYSIioopUUXMgniVlWgciNDQUL774IhITE+Hu7i4qJyIiIjJxZV5Iqnbt2qhdu7aIXIiIiEwCh+YN41LWREREBZjz3AVReI2IiIhIMfZAEBERFSDyYVrmymQKCDn7sdB4UuVqwmLJD9KFxQIAyb6q2Hi2lYXGE070wllWucJiSZZi1x8R/StHbuQjNJ5F7dIt9lZa2rQLQuNJNnbCYmnzxH2fAIBz/Uyh8WBrLzScRaP2QuMJX6DuGcO7MAwzmQKCiIjIVHB83zBeIyIiIlKMPRBEREQFcA6EYSwgiIiICuAcCMM4hEFERESKsQeCiIioAPY/GMYCgoiIqAAVl7I2iEMYREREpFiZeiCysrLw7bff4vz583B1dcWbb76JatUML+Ck0Wig0Wj0N2ZnQ20tdlEfIiIiY/DTtWGKrpG3tzfS0/9ZlfHKlSto2rQpxo0bhz179mDq1Knw9vbGhQuGV6aLjIyEo6OjXota/rVxZ0BERCSYJPA/c6WogDhz5gxyc/9ZHjYiIgJubm64dOkSjh07hkuXLqF58+aYNGmSwTgRERHIyMjQaxPeG2DcGRAREdFTZ/QQRkJCApYtWwZHR0cAgL29PaZPn47Q0FCDr1Wr1VCr1XrbNBy+ICIiE8EhDMMUFxDS/89Mffz4MVxdXfX21apVC7du3RKTGRERUQUx56EHURQXEF26dIGlpSUyMzNx9uxZNG3aVLfv0qVLpZpESUREZMrYA2GYogJi6tSpel/b2+s/jvann36Cn59f2bMiIiIik1amAqKgL774okzJEBERmQI+C8MwrkRJRERUAMsHwzjMQ0RERIqxB4KIiKgADmEYZjIFhKqqW0WnQOVFZSE0nGRbWWg8U2ZR27uiUyiRytmjolMolqpZZ6HxnKLbCY0n2dgJjSfav+nnrCjsnjeM14iIiIgUM5keCCIiIlPBhaQMYwFBRERUALvnDeM1IiIiIsXYA0FERFQABzAMYwFBRERUAG/jNIwFBBERUQEc3zeM14iIiIgUYw8EERFRARzAMExRD0RSUhIuXLig+3r9+vXo0KED6tSpgxdffBGbNm0qVRyNRoPMzEy9ptFolGVORERUTlSQhDVzpaiAePvtt/H3338DAFatWoX33nsPbdq0waRJk/D8889j6NChWLNmjcE4kZGRcHR01Guz5y8z7gyIiIjoqZNkWZZLe3ClSpWQnJwMd3d3tG7dGsOHD8fQoUN1+zdu3IjPP/8cp0+fLjGORqMp1OOgun8NarVaYfpERE+H/DhLaDxTfxaGqbOqXr9c40+q109YrM8vbhQWy5QomgNRqVIl3L59G+7u7rh27Rratm2rt79du3Z6QxzFUavVhYqFnOzbSlIhIiIqN+Y78CCOoiGM7t27Y+nSpQAAf39/fPfdd3r7v/32W3h6eorLjoiIiEySoh6I2bNno0OHDvD390ebNm0wZ84cxMfHw8vLC2fPnsWRI0fwww8/lFeuRERET4U5T34URVEPhJubG06cOAFfX1/ExsZClmUcO3YMu3fvRu3atXH48GH06NGjvHIlIiJ6KlSyuGauFE2iLE85t1MqOgUiomJxEqVpKe9JlNPc+4uLdWmDsFimhAtJERERFcBlmg1jAUFERFQAZ0AYxgKCiIioAPZAGMYCgsiEae+mCo2nsq8iNB4sTPdXSN71s0LjSZWcxMYz9TkQOYIfL2DFhQLNjen+9BMREVUQ3sZpGAsIIiKiAlg+GMZhHiIiIlKMPRBEREQF8NO1YSwgiIiICjDnFSRFYZFFREREirEHgoiIqABOojSMBQQREVEB7J43TNE1Gj16NA4ePFjmN9VoNMjMzNRrGo3gRUuIiIio3CgqIBYvXoxOnTqhYcOGmD17NlJTjVslLzIyEo6Ojnpt9vxlRsUiIiISTSWwmSvF57Z792706NED//nPf1C3bl307t0bMTEx0Gq1pY4RERGBjIwMvTZxzDClqRAREZULlSyumSvFBUSzZs0wb948XL9+HV9//TU0Gg1CQkJQp04dTJo0CefPnzcYQ61Ww8HBQa+p1VwnnYiITIMksJkro3tXrKys8PrrryM2NhYpKSkYOnQoNmzYgEaNGonMj4iIiEyQkOGZunXrYtq0abhw4QJiY2NFhCQiIqownANhmKLbON3d3WFhYVHsfkmS8NJLL5U5KSIioopkzn/4RVF0jS5cuIBq1aqVVy5ERET/eteuXcOAAQNQrVo12NraolmzZvj11191+2VZxpQpU+Dq6gpbW1sEBgbi3LlzejHS09PRv39/ODg4wMnJCUOGDMGDBw/0jjl58iT8/PxgY2ODOnXqICoqSlGeLLKIiIgKkGRxTYm7d++iQ4cOsLKyws6dO/Hnn39izpw5qFKliu6YqKgoLFiwAMuWLcPRo0dhZ2eHoKAgPH78WHdM//79cfr0aezZswcxMTE4cOAA3n33Xd3+zMxMdO3aFe7u7khMTMQXX3yBadOmYcWKFQqukSybxE0mObdTKjoFIpOjvWvcWivFUdlXMXyQEhamu5ht3vWzQuNJlZyExlNVdRMaT7gcwYv7WYm9086qen2h8QpaWXuAsFhDr35d6mM/+ugjHD58uNhFG2VZhpubGz744AN8+OGHAICMjAw4Oztj3bp1CA0NRXJyMry9vXH8+HG0adMGABAbG4sePXrg6tWrcHNzw9KlSzFp0iSkpqbC2tpa997btm3DmTNnSpUreyCIiIjKkZLVl7dv3442bdrgtddeQ82aNdGqVSusXLlSt//ChQtITU1FYGCgbpujoyPatWuHhIQEAEBCQgKcnJx0xQMABAYGQqVS4ejRo7pjOnbsqCseACAoKAhnz57F3bt3S3VepvvxgYigquJS0Sk8syxqe4sNmJcjNp6pE9xj8KwR+ek6MjIS06dP19s2depUTJs2rdCxKSkpWLp0KcaPH4+PP/4Yx48fx/vvvw9ra2sMGjRItwK0s7Oz3uucnZ11+1JTU1GzZk29/ZaWlqhatareMR4eHoVi5O97csikOCwgiIiIChC5gmRERATGjx+vt624xRO1Wi3atGmDmTNnAgBatWqFU6dOYdmyZRg0aJC4pATgEAYREVEBIleiVLL6squrK7y99XvPvLy8cPnyZQCAi8s/vZJpaWl6x6Slpen2ubi44ObNm3r7c3NzkZ6erndMUTGefA9DWEAQERGZiA4dOuDsWf0JwH/99Rfc3d0BAB4eHnBxccG+fft0+zMzM3H06FH4+voCAHx9fXHv3j0kJibqjtm/fz+0Wi3atWunO+bAgQPIyfnf0NyePXvQqFGjUg1fACwgiIiICqmolSjHjRuHI0eOYObMmTh//jw2btyIFStWYOTIkQD+WbBx7Nix+Oyzz7B9+3b88ccfGDhwINzc3BASEgLgnx6Lbt26YejQoTh27BgOHz6MUaNGITQ0FG5u/9z9069fP1hbW2PIkCE4ffo0Nm/ejPnz5xcaaikJ50AQEREVUFFP0Xz++efxww8/ICIiAjNmzICHhwfmzZuH/v37646ZMGECsrKy8O677+LevXt48cUXERsbCxsbG90xGzZswKhRo9ClSxeoVCr07dsXCxYs0O13dHTE7t27MXLkSPj4+KB69eqYMmWK3loRhnAdCCKi0hB9F4aFldh4/zLlvQ7E127i1oEYcL3060A8S9gDQUREVIA5P4ZbFBYQREREBahgEp3zJo2TKImIiEgxxQXEokWLMHDgQGzatAkAsH79enh7e6Nx48b4+OOPkZubazCGkmU9iYiInjaVLK6ZK0UFxGeffYaPP/4YDx8+xLhx4zB79myMGzcO/fv3x6BBg7Bq1Sp8+umnBuNERkbC0dFRr82ev8zokyAiIhKpom7jfJYougvD09MTUVFReOWVV/D777/Dx8cH0dHRuttLfvjhB0yYMKHQc8kL0mg0hXocVPevFbsyFxFRheNdGCalvO/C2OLa3/BBpfTajQ3CYpkSRZMor1+/rnu6V4sWLaBSqdCyZUvd/tatW+P69esG46jV6kLFQk72bSWpEBERlRvehWGYot4VFxcX/PnnnwCAc+fOIS8vT/c1AJw+fbrQE8CIiIieNSpZFtbMlaIeiP79+2PgwIHo3bs39u3bhwkTJuDDDz/EnTt3IEkSPv/8c7z66qvllSsREdFTYc5zF0RRVEBMnz4dtra2SEhIwNChQ/HRRx+hRYsWmDBhAh4+fIhevXqVahIlERERPdu4lDURUWlwEqVJKe9JlD+69BMWq3fqRmGxTAlXoiQiIiqAK1EaxmEeIiIiUow9EERERAWY8wqSophMAaFNN7x+hBKStY3hg0oby76qsFgAxI+larVCw+Vd/dPwQQpY1PYWGk97L1VcrHOJwmIBgEWLzkLj5aUkCY2nqukhNJ6syRIbL1PgejCPHoiLBSB31x6h8Sx79xYaT+X6nNh4VVyExoP0bHV4SxzCMOjZ+hclIiIik2AyPRBERESmgp+uDWMBQUREVAALCMN4jYiIiEgx9kAQEREVwEmUhrGAICIiKoDd84axgCAiIiqAPRCGKS4gbty4gaVLl+LQoUO4ceMGVCoV6tevj5CQEAwePBgWFhblkScRERGZEEW9NL/++iu8vLzw3//+Fzk5OTh37hx8fHxgZ2eHDz/8EB07dsT9+/cNxtFoNMjMzNRrmuxso0+CiIhIJBVkYc1cKSogxo4di3HjxuHXX3/FwYMHsW7dOvz111/YtGkTUlJS8PDhQ0yePNlgnMjISDg6Ouq1qOVfG30SREREIqkkcc1cKSogkpKS8NZbb+m+7tevH5KSkpCWloYqVaogKioK3333ncE4ERERyMjI0GsT3hugPHsiIiKqEIrmQNSsWRM3btxA/fr/PIc9LS0Nubm5cHBwAAA899xzSE9PNxhHrVZDrVbrbdNYWytJhYiIqNxwEqVhinogQkJCMGzYMMTGxiIuLg79+/eHv78/bG1tAQBnz55FrVq1yiVRIiKip0UlsJkrRT0Qn332GW7cuIFevXohLy8Pvr6++Prr/81dkCQJkZGRwpMkIiIi06KogLC3t8fmzZvx+PFj5Obmwt7eXm9/165dhSZHRERUESSJQxiGGLWQlI2Njeg8iIiITIaKBYRB5jw8Q0REROWES1kTEREVYMbLNwjDAoKIiKgADmEYZjIFhFTJUWg8WZMlLJbwStTCSmw8KU9oOIva3kLjwULst5lUubqwWJbtegmLBQCQxI4KWjzXTmg8+WGG0HgWNeoKjafNuCUu1o2/hMUCAFVVB6HxcOu60HDae7eFxpNaBQqNB2tbsfHKGSdRGsY5EERERKSYyfRAEBERmQoOYRjGAoKIiKgAibMoDeIQBhERESnGHggiIqICOInSMKMKiOzsbGzbtg0JCQlITU0FALi4uOCFF15A7969Yc0naxIR0TOMcyAMUzyEcf78eXh5eWHQoEE4ceIEtFottFotTpw4gYEDB6JJkyY4f/58eeRKREREJkJxD8Tw4cPRrFkznDhxAg4O+vdFZ2ZmYuDAgRg5ciR27dolLEkiIqKniZMoDVNcQBw+fBjHjh0rVDwAgIODAz799FO0ayd28RsiIqKniXMgDFNcQDg5OeHixYto2rRpkfsvXrwIJyenEmNoNBpoNBq9bZImG2o1504QERE9CxTPgXjnnXcwcOBAzJ07FydPnkRaWhrS0tJw8uRJzJ07F4MHD8a7775bYozIyEg4OjrqtajFq4w+CSIiIpFUkiysmSvFPRAzZsyAnZ0dvvjiC3zwwQeQ/n+gSJZluLi4YOLEiZgwYUKJMSIiIjB+/Hi9bdKtc0pTISIiKheSynz/8Iti1G2cEydOxMSJE3HhwgW92zg9PDxK9Xq1Wg21Wq23LTuTwxdERGQaOInSsDKtROnh4QFfX1/4+vrqiocrV64gLCxMSHJERERkmoQvZZ2eno7o6GjRYYmIiJ4aSSULa+ZK8RDG9u3bS9yfkpJidDJERESmgLdxGqa4gAgJCYEkSZDl4i+uxMEjIiIis6Z4CMPV1RVbt27VLWFdsCUlJZVHnkRERE8Nb+M0THEB4ePjg8TExGL3G+qdICIiMnWSSlwzV4qHMMLDw5GVlVXsfk9PT8TFxZUpKSIiIjJtigsIPz+/Evfb2dnB39/f6IREkSyMWuLi6dDmCQ0nP8wQGk+yrSw0HvJyhYaTH6SLi2VhJSwWAEhWYtczyTt1QGg8aLVi4zV8Xmg47d8nhMWSankKiwUA6T+lCo1XM2Sg0HhQmfZHXZP+nVwETqI07Nn6FyUiInoKzPn2S1FMu2QlIiIik8QeCCIiogK4GoFhLCCIiIgK4BCGYSwgiIiICmABYZjwORBpaWmYMWOG6LBERERkQoQXEKmpqZg+fbrosERERE+NJIlr5krxEMbJkydL3H/27FmjkyEiIjIFHMIwTHEPRMuWLdGqVSu0bNmyUGvVqhVCQ0MNxtBoNMjMzNRrGk22USdARERkjmbNmgVJkjB27FjdtsePH2PkyJGoVq0a7O3t0bdvX6Slpem97vLlywgODkalSpVQs2ZNhIeHIzdXf0G/+Ph4tG7dGmq1Gp6enli3bp3i/BQXEFWrVsXKlStx4cKFQi0lJQUxMTEGY0RGRsLR0VGvRS1epTh5IiKi8lDRz8I4fvw4li9fjubNm+ttHzduHH766Sds2bIFP//8M65fv45XXnlFtz8vLw/BwcHIzs7GL7/8gujoaKxbtw5TpkzRHXPhwgUEBwcjICAAv/32G8aOHYt33nkHu3btUpSj4iEMHx8fXL9+He7u7kXuv3fvnsGHaUVERGD8+PF626Rb55SmQkREVC4qcinrBw8eoH///li5ciU+++wz3faMjAysXr0aGzduROfOnQEAa9euhZeXF44cOYL27dtj9+7d+PPPP7F37144OzujZcuW+PTTTzFx4kRMmzYN1tbWWLZsGTw8PDBnzhwAgJeXFw4dOoS5c+ciKCio1Hkqro2GDRuGevXqFbu/bt26WLt2bYkx1Go1HBwc9JpaLfYZAkRERKag6GF7TbHHjxw5EsHBwQgMDNTbnpiYiJycHL3tjRs3Rt26dZGQkAAASEhIQLNmzeDs7Kw7JigoCJmZmTh9+rTumIKxg4KCdDFKS3EB0adPHwwYMKDY/VWqVMGgQYOUhiUiIjIZIocwihq2j4yMLPJ9N23ahKSkpCL3p6amwtraGk5OTnrbnZ2dkZqaqjvmyeIhf3/+vpKOyczMxKNHj0p9jYTfxnnlyhWEhYWJDktERPTUSCpZWIuIiEBGRoZei4iIKPSeV65cwZgxY7BhwwbY2NhUwFkrI7yASE9PR3R0tOiwREREz6Sih+3VhY5LTEzEzZs30bp1a1haWsLS0hI///wzFixYAEtLSzg7OyM7Oxv37t3Te11aWhpcXFwAAC4uLoXuysj/2tAxDg4OsLW1LfV5KZ5EuX379hL3p6SkKA1JRERkUoy9e6IsunTpgj/++ENv29tvv43GjRtj4sSJqFOnDqysrLBv3z707dsXwD9rL12+fBm+vr4AAF9fX3z++ee4efMmatasCQDYs2cPHBwc4O3trTvmv//9r9777NmzRxejtBQXECEhIZAkqcQ7LSRzXnqLiIjMXwXchVG5cmU0bdpUb5udnR2qVaum2z5kyBCMHz8eVatWhYODA0aPHg1fX1+0b98eANC1a1d4e3vjrbfeQlRUFFJTUzF58mSMHDlS1+sxbNgwLFq0CBMmTEBYWBj279+Pb7/9Fjt27FCUr+Iay9XVFVu3boVWqy2yJSUlKQ1JRERkUip6HYjizJ07Fz179kTfvn3RsWNHuLi4YOvWrbr9FhYWiImJgYWFBXx9fTFgwAAMHDhQ7xlVHh4e2LFjB/bs2YMWLVpgzpw5WLVqlaJbOAEj14FITExE7969i9xvqHeCiIiISic+Pl7vaxsbGyxevBiLFy8u9jXu7u6FhigK6tSpE06cOFGm3BQXEOHh4cjKyip2v6enJ+Li4sqUFBERUUWqiDkQzxpJNpHugpzbnHxJVEhO8YvNGMWq8MxvKp28lESh8VTVi17N11iSfRWh8aCyEBtPMKvq9cs1/u0gf2Gxqu/6WVgsU8Iai4iIiBRTPIRBRERk9vjx2iAWEERERAVwDoRhvERERESkmNEFxNWrV/HgwYNC23NycnDgwIEyJUVERFShVAKbmVJ8ajdu3EDbtm3h7u4OJycnDBw4UK+QSE9PR0BAgNAkiYiIniZTXUjKlCg+tY8++ggqlQpHjx5FbGws/vzzTwQEBODu3bu6Y0zkzlAiIiIqJ4oLiL1792LBggVo06YNAgMDcfjwYbi6uqJz585IT08HwGdhEBHRM45DGAYpPrWMjAxUqfK/BUvUajW2bt2KevXqISAgADdv3jQYQ6PRIDMzU69pNIIXzCEiIjKSpJKENXOluICoX78+Tp48qbfN0tISW7ZsQf369dGzZ0+DMSIjI+Ho6KjXZs9fpjQVIiKi8sEeCIMUn1r37t2xYsWKQtvzi4iWLVsanAMRERGBjIwMvTZxzDClqRAREVEFUbyQ1Oeff46HDx8WHczSEt9//z2uXbtWYgy1Wq17Lnm+nOzbSlMhIiIqF+Y89CCK4h4IS0tLODg4FLv/xo0bmD59epmSIiIiqlAcwjBI+Kmlp6cjOjpadFgiIiIyIYqHMLZv317i/pQUPpabiIiecRzCMEhxARESEgJJkkqcKMl1IIiI6FnGORCGKR7CcHV1xdatW6HVaotsSUlJ5ZEnERERmRDFBYSPjw8SExOL3W+od4KIiMjkcRKlQYqHMMLDw5GVlVXsfk9PT8TFxZUpKSIiogrFIQyDJNlEugtybnPyJRGZMG2e2HgqC7Hx/mWsqtcv1/gZbwcKi+W4dq+wWKZEcQ8EERGRueMkSsNYQBARERXEAsIgFhBEREQFsYAwyIznhxIREVF5YQ8EERFRAVwQ0TCjCog7d+7g5MmTaNGiBapWrYrbt29j9erV0Gg0eO211+Dl5SU6TyIioqeHQxgGKS4gjh07hq5duyIzMxNOTk7Ys2cPXnvtNVhaWkKr1WLWrFk4dOgQWrduXR75EhERkQlQPAdi0qRJeO2115CRkYGPP/4YISEh6NKlC/766y+cP38eoaGh+PTTT8sjVyIioqdDJYlrZkrxQlJVq1bF4cOH4eXlhZycHNjY2CAhIQFt27YFACQlJeHll1/G1atXi42h0Wig0Wj0tqnuX4NarTbiFIiIngIuJGVSynshqfvv9xQWq/KCGGGxTIniHojs7GzY2toCAKysrFCpUiVUr15dt7969eq4c+dOiTEiIyPh6Oio12bPX6Y0FSIiIqogiudA1KlTBykpKahXrx4AYNOmTXB1ddXtv3Hjhl5BUZSIiAiMHz9eb5vq/jWlqRAREZUPMx56EEVxAREaGoqbN2/qvg4ODtbbv337dt1wRnHUanWh4Yqc7NtKUyEiIioXXMraMOEP03r48CEsLCwUz2fgw7SIyKRxDoRJKe85EA/Gvywslv2X24XFMiXCV6K8c+cOhg8fLjosERHR08O7MAwSXkCkp6cjOjpadFgiIqKnR1KJa2ZK8RyI7dtL7opJSeFQBBERPePMuOdAFMUFREhICCRJQklTJ7iGOBERkXlT3Lfi6uqKrVu3QqvVFtmSkpLKI08iIqKnh3MgDFJcQPj4+CAxMbHY/YZ6J4iIiEydpFIJa+ZK8RBGeHg4srKyit3v6emJuLi4MiVFREREpk1xAeHn51fifjs7O/j7+yvPhPdYExWS+8d+ofFU1eqIjef2nNB42vTr4mJdOyMsFgDhs+kt6jUXGk+yryo0HvJyxMazsBIbr7yZ8dCDKIoLCCIiIrNnxrdfisIrRERERIqxB4KIiKggDmEYxAKCiIioIDO+e0IUYVeofv36OHfunKhwREREZMIU90AsWLCgyO2XL1/G2rVr4eLiAgB4//33y5YZERFRReEQhkGKC4ixY8eiVq1asLTUf6lWq8VXX30FKysrSJLEAoKIiJ5dvAvDIMUFxLvvvoujR49i48aN8PLy0m23srLC7t274e3tLTRBIiKip449EAYpLrGWLVuGKVOmICgoCIsWLTLqTTUaDTIzM/WaRpNtVCwiIiJ6+ozqo+nTpw8SEhLwww8/oHv37khNTVX0+sjISDg6Ouq12QuWGZMKERGRcHwWhmFG38ZZq1Yt7N27F7NmzUKrVq0UPUArIiIC48eP19umyrhibCpERERicQjDoDKtAyFJEiIiItC1a1ccOnQIrq6upXqdWq2GWq3W25bz2LosqRAREdFTJKRvxcfHB2PGjEGVKlVw5coVhIWFiQhLRERUMSSVuGamhJ9Zeno6oqOjRYclIiJ6elSSuGamFA9hbN++vcT9KSkpRidDREREzwbFBURISAgkSSpx0qQkmW/FRURE/wJmfPeEKIqvkKurK7Zu3QqtVltkS0pKKo88iYiInh5JEtfMlOICwsfHB4mJicXuN9Q7QURERM8+xUMY4eHhyMrKKna/p6cn4uLiypQUERFRheIQhkGSbCLdBTm3OfmSiIhKx6p6/XKN/3DxKGGxKo007rEPpq5MC0kRERGZJTO+/VIU9tEQERGRYiwgiIiICqqglSgjIyPx/PPPo3LlyqhZsyZCQkJw9uxZvWMeP36MkSNHolq1arC3t0ffvn2Rlpamd8zly5cRHByMSpUqoWbNmggPD0dubq7eMfHx8WjdujXUajU8PT2xbt06RbmygCAiIipIpRLXFPj5558xcuRIHDlyBHv27EFOTg66du2qd/PCuHHj8NNPP2HLli34+eefcf36dbzyyiu6/Xl5eQgODkZ2djZ++eUXREdHY926dZgyZYrumAsXLiA4OBgBAQH47bffMHbsWLzzzjvYtWtXqXPlJEoiInrmlPskyhXjhMWq9O5co19769Yt1KxZEz///DM6duyIjIwM1KhRAxs3bsSrr74KADhz5gy8vLyQkJCA9u3bY+fOnejZsyeuX78OZ2dnAMCyZcswceJE3Lp1C9bW1pg4cSJ27NiBU6dO6d4rNDQU9+7dQ2xsbKlyK3MPhCzLiIuLw8qVKxETE4OcnJyyhiQiIqpQkkoS1jQaDTIzM/WaRqMpVR4ZGRkAgKpVqwIAEhMTkZOTg8DAQN0xjRs3Rt26dZGQkAAASEhIQLNmzXTFAwAEBQUhMzMTp0+f1h3zZIz8Y/JjlIbiAqJHjx66E0pPT4evry+6dOmCSZMmoXfv3mjevDlu3bqlNCwREZHpEDgHIjIyEo6OjnotMjLSYAparRZjx45Fhw4d0LRpUwBAamoqrK2t4eTkpHess7MzUlNTdcc8WTzk78/fV9IxmZmZePToUakukeICIjY2Vlc5TZ48Gffv38fff/+Nmzdv4tKlS7Czs9MbZyEiIvo3i4iIQEZGhl6LiIgw+LqRI0fi1KlT2LRp01PIUrkyrQOxf/9+REVFwcPDAwBQu3ZtzJ49G0OHDi3xdRqNplD3jUqjgVqtLks6REREYghciVKtViv++zZq1CjExMTgwIEDqF27tm67i4sLsrOzce/ePb1eiLS0NLi4uOiOOXbsmF68/Ls0njym4J0baWlpcHBwgK2tbalyNOoK5T9t8+7du2jQoIHePk9PT1y/fr3E1xfVnTN7/jJjUiEiIhKvgu7CkGUZo0aNwg8//ID9+/frPqDn8/HxgZWVFfbt26fbdvbsWVy+fBm+vr4AAF9fX/zxxx+4efOm7pg9e/bAwcEB3t7eumOejJF/TH6M0jCqB2Lw4MFQq9XIycnBhQsX0KRJE92+1NTUQmMzBUVERGD8+PF621T3rxmTChERkdkYOXIkNm7ciB9//BGVK1fWzVlwdHSEra0tHB0dMWTIEIwfPx5Vq1aFg4MDRo8eDV9fX7Rv3x4A0LVrV3h7e+Ott95CVFQUUlNTMXnyZIwcOVLXEzJs2DAsWrQIEyZMQFhYGPbv349vv/0WO3bsKHWuiguIQYMG6f6/d+/eePjwod7+77//Hi1btiwxRlHdOTnZt5WmQkREVD4q6DHcS5cuBQB06tRJb/vatWsxePBgAMDcuXOhUqnQt29faDQaBAUFYcmSJbpjLSwsEBMTg+HDh8PX1xd2dnYYNGgQZsyYoTvGw8MDO3bswLhx4zB//nzUrl0bq1atQlBQUKlzFb4ORFZWFiwsLGBjY6PodVwHgoiISqu814F49PUkYbFsB3wuLJYpEb4SZXp6OkaMGCE6LBER0dNTQUtZP0vKpYCIjo4WHZaIiIhMiOI5ENu3by9xf0oKhyKIiOgZJ/A2TnOluIAICQmBJEkoaeqEVEGTT4iIiIRgAWGQ4ivk6uqKrVu3QqvVFtmSkpLKI08iIiIyIYoLCB8fHyQmJha731DvBBERkcmTJHHNTCkewggPD9d7LnlBnp6eiIuLK1NSREREFYpDGAYpLiD8/PxK3G9nZwd/f3/FiciZYheSkmwrC4sl52YLiwWIza1caPPExlNZCA0nP0gXF8zCSlwsAJK1svVPDMm7miw0HiythYazqNVYaDztrUvCYuXt2CAsFgAk/uee0HjPb+ktNJ6qViOh8WBjJzScZFGmRy+RCeK/KBERUUHsgTCIBQQREVFBZrwAlCi8QkRERKQYeyCIiIgK4hCGQYqv0NWrV3H79v8mPB48eBD9+/eHn58fBgwYgISEBKEJEhERPXV8FoZBis+sb9++OHLkCADgxx9/RKdOnfDgwQN06NABDx8+hL+/P2JiYoQnSkRE9NSoVOKamVI8hHH69Gk0adIEABAZGYmZM2di4sSJuv2LFi3ClClT0LNnT3FZEhERkUlRXBpZWlri/v37AIALFy6ge/fuevu7d++Os2fPismOiIioInAIwyDFZ+bv749vvvkGANCqVSvEx8fr7Y+Li0OtWrVKjKHRaJCZmanXNBqxizUREREZjUMYBikewpg1axb8/Pxw/fp1vPjii5g0aRKOHz8OLy8vnD17Fps3b8ayZctKjBEZGYnp06frbZv8/jv4ZMxQpekQERFRBVBcQHh5eeHo0aOYPHkyoqKikJWVhQ0bNsDS0hLPP/88Nm3ahJCQkBJjREREYPz48XrbpGsnlaZCRERUPsy450AUo9aBaNCgAb755hvIsoybN29Cq9WievXqsLIq3XMF1Go11Gq13rbs22LX6CciIjKaGc9dEKVMV0iSJDg7O8PV1VVXPFy5cgVhYWFCkiMiIiLTJLzESk9PR3R0tOiwRERETw8nURqkeAhj+/btJe5PSUkxOhkiIiKTwCEMgxQXECEhIZAkCbIsF3uMJEllSoqIiIhMm+ISy9XVFVu3boVWqy2yJSUllUeeRERETw+HMAxSfGY+Pj5ITEwsdr+h3gkiIiKTx5UoDVI8hBEeHo6srKxi93t6eiIuLq5MSREREVUoM+45EEVxAeHn51fifjs7O/j7+ytORH50X/FrSmRh1BIXRdLeSxUWCwAsajUWGg/aPKHh5Ad3hcaDjZ3YeBalW2+kVARfO9Hkk78IjSc1bSc0nvau2J8N+UG6sFhSk+bCYgFAmzm3hMaTH2YIjSc5VBcaT/TvZDkvV2g8qnji/soSERGZCUmyqOgUTB4LCCIiooI4hGEQrxAREREpxh4IIiKigtgDYRALCCIiooLM+PZLURRfoTlz5uDSpUvlkQsRERE9IxQXEOHh4WjQoAFeeuklbN68GdnZ2eWRFxERUcXhSpQGGXVmq1atgp2dHd566y24ublh7NixOHXqlOjciIiIKgZXojTIqDPr0aMHtm3bhqtXr2LChAnYtWsXWrRogbZt22LlypW4f7/kBUg0Gg0yMzP1moY9GURERM+MMpVGNWvWxIQJE5CcnIz4+Hh4e3tj3LhxcHV1LfF1kZGRcHR01GtRK78pSypERETiqCzENTOl+C6M4h7V7efnBz8/PyxYsACbN28uMUZERATGjx+vv/Hvw0pTISIiKh9mPPQgiuICwtCTNh0cHDB06NASj1Gr1VCr1XrbNNbWSlMhIiIqH2Y8+VEUxQWEVqstjzyIiIjoGSK8xLpy5QrCwsJEhyUiInp6eBeGQcLPLD09HdHR0aLDEhERPTWSZCGsmSvFQxjbt28vcX9KSorRyRAREdGzQXEBERISAkmSSpxMWdydGkRERM8ETqI0SPEVcnV1xdatW6HVaotsSUlJ5ZEnERHR08M5EAYpPjMfHx8kJiYWu99Q7wQRERE9+xQPYYSHhyMrK6vY/Z6enoiLiytTUkRERBWKQxgGKS4g/Pz8StxvZ2cHf39/xYlItpUVv6ZEFlbiQtVqLCxWuRC8VKrwfwsrteFjFJCz7okLJjg3kd93AKDyDRYaT/T3isqxhtB4srWNsFh5f/0mLBYAHP/gjNB4zy8X+3MmPyr5GUSK4z1IFxpPcqwpNF65M+OhB1F4hYiIiEgxxT0QREREZs+MH4IlCgsIIiKigjiEYRALCCIiooI4idIgXiEiIiJSzKgCIiYmBlOmTMHhw4cBAPv370ePHj3QrVs3rFixQmiCRERET5skqYQ1c6X4zJYvX44+ffrgv//9L3r06IGvv/4aISEhqFWrFurVq4exY8di/vz55ZErERHR06FSiWtmSvEciAULFmDJkiUYOnQo4uLi0KNHD8yZMwcjRowAALRv3x5RUVEYM2aM8GSJiIjINCgujS5cuICgoCAAQEBAAPLy8tCxY0fd/k6dOuHSpUslxtBoNMjMzNRrGk220lSIiIjKB5+FYZDiM6tWrZquQLh+/Tpyc3Nx+fJl3f5Lly6hatWqJcaIjIyEo6OjXotaFq00FSIiovKhshDXzJTiIYzevXtjyJAhGDRoELZv346BAwfigw8+gEqlgiRJCA8PR9euXUuMERERgfHjx+ttk66dVJoKERERVRDFBcTs2bORnZ2NTZs24YUXXsDChQuxYMEC9O7dGzk5OfD390dkZGSJMdRqNdRq/WcQZN+2VpoKERFR+TDjoQdRFBcQdnZ2hW7V/PDDDzFq1Cjk5OSgcmXBD2IiIiJ62sz47glRhF0hGxsbVK5cGVeuXEFYWJiosERERGSChJdY6enpiI7mhEgiInp2cSEpwxQPYWzfvr3E/SkpKUYnQ0REZBIqcAhj8eLF+OKLL5CamooWLVpg4cKFaNu2bYXlUxzFBURISAgkSYIsy8UeI0lSmZIiIiKqUBXUc7B582aMHz8ey5YtQ7t27TBv3jwEBQXh7NmzqFmzZoXkVBzFV8jV1RVbt26FVqstsiUlJZVHnkRERGbvyy+/xNChQ/H222/D29sby5YtQ6VKlbBmzZqKTq0QxQWEj48PEhMTi91vqHeCiIjI5AlcSKro1Zc1hd4yOzsbiYmJCAwM/F8aKhUCAwORkJDwNM++dGSFDhw4IO/cubPY/Q8ePJDj4+OVhi2Vx48fy1OnTpUfP35sUrEYz7zjmXJujGc6sRjP9OKZiqlTp8oA9NrUqVMLHXft2jUZgPzLL7/obQ8PD5fbtm37lLItPcUFREXKyMiQAcgZGRkmFYvxzDueKefGeKYTi/FML56pePz4sZyRkaHXiiqSnrUCQvEkSiIiIiq9olZfLkr16tVhYWGBtLQ0ve1paWlwcXEpr/SMZr43qBIRET1DrK2t4ePjg3379um2abVa7Nu3D76+vhWYWdHYA0FERGQixo8fj0GDBqFNmzZo27Yt5s2bh6ysLLz99tsVnVohz1QBoVarMXXq1FJ1BT3NWIxn3vFMOTfGM51YjGd68Z5Fb7zxBm7duoUpU6YgNTUVLVu2RGxsLJydnSs6tUIkWeY9l0RERKQM50AQERGRYiwgiIiISDEWEERERKQYCwgiIiJSjAUEERERKWbyBcTgwYMREhJSaHt8fDwkScK9e/fKHO+7776DjY0N5syZY1R+kiRBkiRYWVnBw8MDEyZMwOPHj8scy9nZGS+99BLWrFkDrVarOF7BmE+2bt26GR1r1qxZetu3bdtm9CPcr1y5grCwMLi5ucHa2hru7u4YM2YM7ty5Y1S8W7duYfjw4ahbty7UajVcXFwQFBSEw4cPK4pT1DV7sk2bNs2o/AAgISEBFhYWCA4ONjrGk4r7GamoOL169Sr2++vgwYOQJAknT55UlJckSRg2bFihfSNHjoQkSRg8eLDiPIv72Th//nypXr9s2TJUrlwZubm5um0PHjyAlZUVOnXqpHds/u+rv//+22BcWZYRGBiIoKCgQvuWLFkCJycnXL16tVQ55ktNTcWYMWPg6ekJGxsbODs7o0OHDli6dCkePnyoKNaTMUePHo369etDrVajTp066NWrl94iSKX15L+FtbU1PD09MWPGDL1rS6bH5AuI8rZq1Sr0798fS5cuxQcffGBUjG7duuHGjRtISUnB3LlzsXz5ckydOrVMsS5evIidO3ciICAAY8aMQc+ePY3+YcqP+WT75ptvjIplY2OD2bNn4+7du0a9/kkpKSlo06YNzp07h2+++Qbnz5/HsmXLdKuupaenK47Zt29fnDhxAtHR0fjrr7+wfft2dOrUSXFB8uS1mjdvHhwcHPS2ffjhh4pzy7d69WqMHj0aBw4cwPXr142OY6qGDBmCPXv2FPlHbu3atWjTpg2aN2+uKGadOnWwadMmPHr0SLft8ePH2LhxI+rWrWt0rkX9bHh4eJTqtQEBAXjw4AF+/fVX3baDBw/CxcUFR48e1fsQERcXh7p166JBgwYG40qShLVr1+Lo0aNYvny5bvuFCxcwYcIELFy4ELVr1y71OaakpKBVq1bYvXs3Zs6ciRMnTiAhIQETJkxATEwM9u7dW+pY+S5evAgfHx/s378fX3zxBf744w/ExsYiICAAI0eOVBwP+N+/xblz5/DBBx9g2rRp+OKLL4yKRU/HM7WQlGhRUVGYOnUqNm3ahD59+hgdJ/+TLvDPL7rAwEDs2bMHs2fPLlOsWrVqoXXr1mjfvj26dOmCdevW4Z133ilTzLIKDAzE+fPnERkZiaioqDLFGjlyJKytrbF7927Y2toCAOrWrYtWrVqhQYMGmDRpEpYuXVrqePfu3cPBgwcRHx8Pf39/AIC7uzvatm2rOLcnr5ejoyMkSRJyDR88eIDNmzfj119/RWpqKtatW4ePP/64zHFNSc+ePVGjRg2sW7cOkydP1m1/8OABtmzZYtQfhdatW+Pvv//G1q1b0b9/fwDA1q1bUbdu3VL/wS9KWX42GjVqBFdXV8THx6N9+/YA/ulp6N27N/bv348jR47oeiLi4+MREBBQ6th16tTB/PnzMWrUKHTt2hX16tXDkCFD0LVrV7z11luK8hwxYgQsLS3x66+/ws7OTre9fv366N27N4xZCmjEiBGQJAnHjh3Ti9mkSROEhYUpjgfo/1sMHz4cP/zwA7Zv346IiAij4lH5+9f2QEycOBGffvopYmJiylQ8FHTq1Cn88ssvsLa2Fhazc+fOaNGiBbZu3SosprEsLCwwc+ZMLFy4UHE36pPS09Oxa9cujBgxQlc85HNxcUH//v2xefNmRb/c7O3tYW9vj23btkGj0RidW3n69ttv0bhxYzRq1AgDBgzAmjVrjPoFbsosLS0xcOBArFu3Tu/ctmzZgry8PLz55ptGxQ0LC8PatWt1X69Zs6bCl/cNCAhAXFyc7uu4uDh06tQJ/v7+uu2PHj3C0aNHFRUQADBo0CB06dIFYWFhWLRoEU6dOqXXI1Ead+7cwe7duzFy5Ei9P/RPUjr8mJ6ejtjY2GJjOjk5KYpXHFtbW2RnZwuJReXjmSggYmJidH8c8lv37t2Njrdz505ERUXhxx9/RJcuXYTlZ2Njg2bNmuHmzZsIDw8vc9wnNW7cGBcvXixTfk+2mTNnGp1Lnz590LJlS6OHaQDg3LlzkGUZXl5eRe738vLC3bt3cevWrVLHtLS0xLp16xAdHQ0nJyd06NABH3/8saLx9vK2evVqDBgwAMA/XbYZGRn4+eefKzgr8cLCwvD333/rndvatWvRt29fODo6GhVzwIABOHToEC5duoRLly7h8OHDumtprII/G6+99pqi1wcEBODw4cPIzc3F/fv3ceLECfj7+6Njx46Ij48H8M+cF41Go7iAAIAVK1bg1KlTGDt2LFasWIEaNWooev358+chyzIaNWqkt7169eq6c544caJRMRs3bqzodaUlyzL27t2LXbt2oXPnzuXyHiTGMzGEERAQUKgr++jRo0b/8mjevDlu376NqVOnom3btrC3txeSX1ZWFubOnQtLS0v07du3TDELkmXZ6ImKRV2/qlWrlimf2bNno3PnzmWaCwDA4KdvpT05ffv2RXBwMA4ePIgjR47oisVVq1YZNdFOpLNnz+LYsWP44YcfAPxT8LzxxhtYvXp1oUl3z7rGjRvjhRdewJo1a9CpUyecP38eBw8exIwZM4yOWaNGDQQHB+t6NoKDg1G9evUy5VnwZ6O4T+nF6dSpE7KysnD8+HHcvXsXDRs2RI0aNeDv74+3334bjx8/Rnx8POrXr2/UXI2aNWvivffew7Zt24RMcM137NgxaLVa9O/fX3FvXXn1mOUXczk5OdBqtejXr1+ZJitT+XsmCgg7Ozt4enrqbStL93mtWrXw3XffISAgAN26dcPOnTtRuXJlIfmtWbMGLVq0wOrVqzFkyBCjYxaUnJxs9FhvUdevrDp27IigoCBEREQY9YfZ09MTkiQhOTm5yCGk5ORk1KhRw6juUBsbG7z00kt46aWX8Mknn+Cdd97B1KlTK7yAWL16NXJzc+Hm5qbbJssy1Go1Fi1aZPQnc1M1ZMgQjB49GosXL8batWvRoEED3dwUY4WFhWHUqFEAgMWLF5c5x7L+bHh6eqJ27dqIi4vD3bt3defn5uaGOnXq4JdffkFcXFyZPklbWlrC0tK4X9X5P2dnz57V216/fn0AKDR8WBrPPfccJEnCmTNnjMqpOPnFnLW1Ndzc3Iw+Z3p6nokhjPLg7u6On3/+GampqejWrRvu378vJK5KpcLHH3+MyZMn680YL4v9+/fjjz/+EN6rUVazZs3CTz/9hISEBMWvrVatGl566SUsWbKk0HVKTU3Fhg0bhP3B9/b2RlZWlpBYxsrNzcVXX32FOXPm4LffftO133//HW5ubkbfFWPKXn/9dahUKmzcuBFfffUVwsLCjO5Fy9etWzdkZ2cjJyenyNscK0JAQADi4+MRHx+v15PUsWNH7Ny5E8eOHTNq+EKE/J+zRYsWCfsZqFq1KoKCgrB48eIiYyq9tT5ffjFXt25dFg/PiH9tAQH8M9M5Pj4eN2/eRFBQEDIzM4XEfe2112BhYWHUJySNRoPU1FRcu3YNSUlJmDlzJnr37o2ePXti4MCBRuWTH/PJdvv2baNiPalZs2bo378/FixYYNTrFy1aBI1Gg6CgIBw4cABXrlxBbGwsXnrpJTRs2BBTpkxRFO/OnTvo3Lkzvv76a5w8eRIXLlzAli1bEBUVhd69exuVoygxMTG4e/cuhgwZgqZNm+q1vn37YvXq1RWaX76MjAy9Aue3337DlStXjIplb2+PN954AxEREbhx44aQgtDCwgLJycn4888/YWFhUeZ4IgQEBODQoUP47bff9HpY/P39sXz5cmRnZ1dYAQH8s3ZEbm4u2rRpg82bNyM5ORlnz57F119/jTNnzhh1HRcvXoy8vDy0bdsW33//Pc6dO4fk5GQsWLAAvr6+5XAWZIr+1QUEANSuXRvx8fG4ffu2sCLC0tISo0aNQlRUlOKqPzY2Fq6urqhXrx66deuGuLg4LFiwAD/++KPRvzDzYz7ZXnzxRaNiFTRjxgyjF7l67rnncPz4cdSvXx+vv/463N3d0b17dzRs2BCHDx9WPDfF3t4e7dq1w9y5c9GxY0c0bdoUn3zyCYYOHYpFixYZlaMoq1evRmBgYJHDFH379sWvv/5qEpM94+Pj0apVK702ffp0o+MNGTIEd+/eRVBQkN7QTVk4ODjAwcFBSCwRAgIC8OjRI3h6esLZ2Vm33d/fH/fv39fd7llRGjRogBMnTiAwMBARERFo0aIF2rRpg4ULF+LDDz/Ep59+qjhm/fr1kZSUhICAAHzwwQdo2rQpXnrpJezbt0/Rrdf0bJNkc7uHjJ5pU6dOxZdffok9e/bo7q0nIiLTwwKCTM7atWuRkZGB999/HyrVv76TjIjIJLGAICIiIsX48Y6IiIgUYwFBREREirGAICIiIsVYQBAREZFiLCCIiIhIMRYQREREpBgLCCIiIlKMBQQREREpxgKCiIiIFPs/r6DXR4ehljwAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# prepare the matrix to plot the heatmap\n",
    "mamm[\"difference_in_seq\"] = [diff_func(x) for x in mamm[\"Sequence\"]]\n",
    "diff = mamm[\"difference_in_seq\"].str.split(\"\", n= -1, expand = True)\n",
    "mamm =mamm.merge(diff, left_index=True, right_index=True)\n",
    "\n",
    "\n",
    "# charged->polar->aliphatic->aromatic->other\n",
    "aa_list = ['H', 'K', 'R', 'D', 'E', 'N', 'Q', 'S', 'T', 'A', 'I', 'L', 'V', 'M', 'F', 'W', 'Y','G', 'C', 'P']\n",
    "\n",
    "#aa_list = ['H', 'K', 'R', 'D', 'E', 'A', 'I', 'L', 'V', 'M', 'F', 'W', 'Y', 'N', 'Q', 'S', 'T', 'G', 'C', 'P']\n",
    "mtx = pd.DataFrame(0, index=aa_list, columns=np.arange(1, 19))\n",
    "\n",
    "# fill the matix with the count of each amino acid in each position\n",
    "for i in range(1, 19):\n",
    "    pos_count = mamm[i].value_counts().to_dict()\n",
    "\n",
    "    for k,v in pos_count.items():\n",
    "        if k != '-':\n",
    "            mtx.loc[k,i] = v\n",
    "\n",
    "# plot the heatmap\n",
    "plt.title(\"Mammalian Specific Peptides\")\n",
    "sns.heatmap(mtx.T, cmap=\"rocket_r\")\n",
    "#plt.savefig('results/changes_heatmap_predictions_mamm.png' , dpi=1200)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "bac.to_excel('results/bac_specific_peptides.xlsx')\n",
    "mamm.to_excel('results/mamm_specific_peptides.xlsx')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "ML",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}