-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathall.py
208 lines (176 loc) · 11.4 KB
/
all.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
#!/usr/bin/env python
# coding: utf-8
import re
import os
import sys
import pandas as pd
import joblib
import streamlit as st
from all_rule import RULE_CHECK
from rdkit.Chem import MolFromSmiles, MolToSmiles, AllChem, MACCSkeys, RDKFingerprint, Lipinski
from rdkit.Chem.Descriptors import CalcMolDescriptors
class Molecule:
def __init__(self, smiles: str):
if not smiles :
print("Empty smiles are given")
sys.exit()
self.smiles = smiles
self.mol = MolFromSmiles(smiles)
def descriptor_generator(self):
return CalcMolDescriptors(self.mol)
def fingerprint_generator(self, fp_type: str, n_bits: int=2048):
fingerprint_type = {
'Morgan': AllChem.GetMorganFingerprintAsBitVect,
'RDKit': RDKFingerprint,
'Atom': AllChem.GetHashedAtomPairFingerprintAsBitVect,
'MACCS': MACCSkeys.GenMACCSKeys,
'Topological': AllChem.GetHashedTopologicalTorsionFingerprintAsBitVect,
'ECFP2': AllChem.GetMorganFingerprintAsBitVect,
'ECFP4': AllChem.GetMorganFingerprintAsBitVect
}
if fp_type not in fingerprint_type:
print("Unknown Fingerprint")
sys.exit()
if n_bits < 1:
print("Number of bits can't be less than 1")
sys.exit()
fingerprint_function = fingerprint_type[fp_type]
if fp_type == 'Morgan':
fingerprint= fingerprint_function(self.mol, 2, nBits=n_bits)
elif fp_type == 'ECFP2':
fingerprint= fingerprint_function(self.mol, 1, nBits=n_bits)
elif fp_type == 'MACCS':
fingerprint = fingerprint_function(self.mol)
else:
fingerprint = fingerprint_function(self.mol, nBits=n_bits)
fingerprint_list = fingerprint.ToList()
fingerprint_dict = {f'{i}': int(bit) for i, bit in enumerate(fingerprint_list)}
return fingerprint_dict
def convert_to_molecule(smiles: str):
try:
return Molecule(smiles)
except ValueError:
return None
def convert_to_check_rule(smiles: str):
try:
return RULE_CHECK(smiles)
except ValueError:
return None
SMI = st.text_input('Input SMILE', 'O=Cc1ccc(Cl)cc1')
st.write('The input SMILE is', str(SMI))
mol = convert_to_molecule(SMI)
mol2 = convert_to_check_rule(SMI)
descriptors = mol.descriptor_generator()
descriptors_1 = pd.DataFrame([list(descriptors.values())], columns= list(descriptors.keys()))
MACCS_fingerprint = mol.fingerprint_generator('MACCS', 2048)
ECFP4_fingerprint = mol.fingerprint_generator('Morgan', 1024)
ECFP2_fingerprint = mol.fingerprint_generator('ECFP2', 1024)
Morgan_fingerprints_dataframe = mol.fingerprint_generator("Morgan", 2048)
MACCS_fingerprint_1 = pd.DataFrame([list(MACCS_fingerprint.values())], columns= list(MACCS_fingerprint.keys()))
ECFP4_fingerprint_1 = pd.DataFrame([list(ECFP4_fingerprint.values())], columns= list(ECFP4_fingerprint.keys()))
ECFP2_fingerprint_1 = pd.DataFrame([list(ECFP2_fingerprint.values())], columns= list(ECFP2_fingerprint.keys()))
Morgan_fingerprints_dataframe = pd.DataFrame([list(Morgan_fingerprints_dataframe.values())], columns= list(Morgan_fingerprints_dataframe.keys()))
descriptors_without_IPC = descriptors_1.drop(['Ipc'], axis=1)
desc_with_fp = pd.merge(descriptors_1,Morgan_fingerprints_dataframe, how='left', left_index=True, right_index=True)
descriptors_without_IPC2 = descriptors_without_IPC
descriptors_without_IPC2 = descriptors_without_IPC2.reset_index()
descriptors_without_IPC2['index'] = descriptors_without_IPC2.pop('index')
st.markdown(''':rainbow[PHYSICOCHEMICAL]''')
# st.write("PHYSICOCHEMICAL")
# st.write("HeavyAtomCount : ", mol.lipinski())
# st.write("HeavyAtomCount: ", round(descriptors_1['HeavyAtomCount'][0],4))
list_descriptors = ['MolWt','MolLogP','MolMR','qed','HeavyAtomMolWt','ExactMolWt','NumValenceElectrons','NumRadicalElectrons','MaxPartialCharge','MinPartialCharge','MaxAbsPartialCharge','MinAbsPartialCharge','FpDensityMorgan1','FpDensityMorgan2','FpDensityMorgan3','TPSA','FractionCSP3','HeavyAtomCount','NHOHCount','NOCount','NumAliphaticCarbocycles','NumAliphaticHeterocycles','NumAliphaticRings','NumAromaticCarbocycles','NumAromaticHeterocycles','NumAromaticRings','NumHAcceptors','NumHDonors','NumHeteroatoms','NumRotatableBonds','NumSaturatedCarbocycles','NumSaturatedHeterocycles','NumSaturatedRings','RingCount']
for desc in list_descriptors:
st.write(desc,':', round(descriptors_1[desc][0],4))
st.markdown(''':rainbow[Different Rule Check]''')
# st.write("Rule Check:")
st.write("Lipinski Rule Violation Check: ", mol2.druglikeness_lipinski())
st.write("Conditions for failure : h_bond_donors > 5, h_bond_acceptors > 10, molecular_weight > 500 and logp > 5 ")
st.write("Egan Rule Violation Check: ", mol2.druglikeness_egan())
st.write("Conditions for failure : tpsa > 131.6 and logp > 5.88 ")
st.write("Ghose Rule Violation Check: ", mol2.druglikeness_ghose())
st.write("Conditions for failure : logp > 5.6 or logp < -0.4 ,molar_refractivity < 40 or molar_refractivity > 130, molecular_weight < 160 or molecular_weight > 480 and n_atoms < 20 or n_atoms > 70")
st.write("Ghose Preferred Rule Violation Check: ", mol2.druglikeness_ghose_pref())
st.write("Conditions for failure : logp > 4.1 or logp < 1.3, molecular_weight < 230 or molecular_weight > 390, molar_refractivity < 70 or molar_refractivity > 110 and n_atoms < 30 or n_atoms > 55 ")
st.write("Muegge Rule Violation Check: ", mol2.druglikeness_muegge())
st.write("Conditions for failure : molecular_weight > 600 or molecular_weight < 200, logp > 5 or logp < -2, tpsa > 150, n_rings > 7, n_carbon < 5, n_heteroatoms < 2, n_rot_bonds > 15, h_bond_acc > 10 and h_bond_don > 5 ")
st.write("Veber Rule Violation Check: ", mol2.druglikeness_veber())
st.write("Conditions for failure : tpsa > 140 and n_rot_bonds > 10")
st.write(f"PAINS filter: {mol2.pains()}")
st.write(f"Brenk filter: {mol2.brenk()}")
# Absorption
print("Absorption loading .....")
# absorption_caco2_rf = joblib.load('model_saved/absorption_caco2_rf_model.joblib')
# absorption_HIA_rf = joblib.load('model_saved/absorption_HIA_rf_model.joblib')
# absorption_Bioavailability_Ma_MACCS_rf = joblib.load('model_saved/absorption_Bioavailability_Ma_MACCS_rf_model.joblib')
# Lipophilicity_AstraZeneca_MACCS_rf = joblib.load('model_saved/Lipophilicity_AstraZeneca_MACCS_rf_model.joblib')
# Pgp_Broccatelli_inhibitor_ECFP4 = joblib.load('model_saved/Pgp_Broccatelli_inhibitor_ECFP4_fingerprint.joblib')
st.markdown(''':rainbow[ABSORPTION]''')
# st.write("ABSORPTION")
st.write("Absorption_Caco2 : ", "Removed due to Github LFS policy")#round(absorption_caco2_rf.predict(descriptors_1)[0],4))
st.write("Absorption_HIA : ", "Removed due to Github LFS policy")# round(absorption_HIA_rf.predict(MACCS_fingerprint_1)[0],4))
st.write("Absorption_Bioavailability : ", "Removed due to Github LFS policy")# round(absorption_Bioavailability_Ma_MACCS_rf.predict(MACCS_fingerprint_1)[0],4))
st.write("Absorption_Lipophilicity : ", "Removed due to Github LFS policy")#round(Lipophilicity_AstraZeneca_MACCS_rf.predict(MACCS_fingerprint_1)[0],4))
st.write("Absorption_PgpInhibotor : ", "Removed due to Github LFS policy")# round(Pgp_Broccatelli_inhibitor_ECFP4.predict(ECFP4_fingerprint_1)[0],4))
# Distribution
print("Distribution loading .....")
# VDss_Lombardo_rf = joblib.load('model_saved/VDss_Lombardo_rf_model.joblib')
# PPBR_AZ_rf = joblib.load('model_saved/PPBR_AZ_rf_model.joblib')
# BBB_Martins_ECFP2_fingerprint = joblib.load('model_saved/BBB_Martins_ECFP2_fingerprint.joblib')
st.markdown(''':rainbow[DISTRIBUTION]''')
# st.write("DISTRIBUTION")
st.write("VDss : ", "Removed due to Github LFS policy")# round(VDss_Lombardo_rf.predict(descriptors_without_IPC)[0],4))
st.write("PPBR : ", "Removed due to Github LFS policy")# round(PPBR_AZ_rf.predict(descriptors_1)[0],4))
st.write("BBB_Martins : ", "Removed due to Github LFS policy")# round(BBB_Martins_ECFP2_fingerprint.predict(ECFP2_fingerprint_1)[0],4))
# Metabolism
print("Metabolism loading .....")
# CYP1A2_random_forest = joblib.load('model_saved/CYP1A2_random_forest_model.joblib')
# CYP2C9_random_forest = joblib.load('model_saved/CYP2C9_random_forest_model.joblib')
# CYP2C19_random_forest = joblib.load('model_saved/CYP2C19_random_forest_model.joblib')
# CYP2D6_random_forest = joblib.load('model_saved/CYP2D6_random_forest_model.joblib')
# CYP3A4_random_forest = joblib.load('model_saved/CYP3A4_random_forest_model.joblib')
st.markdown(''':rainbow[METABOLISM]''')
# st.write("METABOLISM")
st.write("CYP1A2 : ", "Removed due to Github LFS policy")# round(CYP1A2_random_forest.predict(desc_with_fp)[0],4))
st.write("CYP2C9 : ", "Removed due to Github LFS policy")# round(CYP2C9_random_forest.predict(desc_with_fp)[0],4))
st.write("CYP2C19 : ", "Removed due to Github LFS policy")# round(CYP2C19_random_forest.predict(desc_with_fp)[0],4))
st.write("CYP2D6 : ", "Removed due to Github LFS policy")# round(CYP2D6_random_forest.predict(desc_with_fp)[0],4))
st.write("CYP3A4 : ", "Removed due to Github LFS policy")# round(CYP3A4_random_forest.predict(desc_with_fp)[0],4))
# Excretion
print("Excretion loading .....")
# half_life = joblib.load('model_saved/excretion_Half_Life_Obach_descriptor_RF_model.joblib')
# clearance = joblib.load('model_saved/Clearance_Hepatocyte_AZ_rf_model.joblib')
st.markdown(''':rainbow[EXCREATION]''')
# st.write("EXCREATION")
st.write("T 1/2 half life", "Removed due to Github LFS policy")#round(half_life.predict(descriptors_without_IPC)[0],4))
st.write("Clearance ", "Removed due to Github LFS policy")# round(clearance.predict(descriptors_1)[0],4))
# Toxicity
print("Toxicity loading .....")
# AMES_MACCS = joblib.load('model_saved/AMES_MACCS_rf_model.joblib')
# Carcinogens_Lagunin = joblib.load('model_saved/Carcinogens_Lagunin_MACCS_rf_model.joblib')
# ClinTox_MACCS = joblib.load('model_saved/ClinTox_MACCS_rf_model.joblib')
# DILI_MACCS = joblib.load('model_saved/DILI_MACCS_rf_model.joblib')
# hERG_tox = joblib.load('model_saved/hERG_tox_rf_model.joblib')
# LD50 = joblib.load('model_saved/LD50_Zhu_tox_rf_model.joblib')
# Skin_Reaction = joblib.load('model_saved/Skin_Reaction_MACCS_rf_model.joblib')
## Tox 21
print("Tox 21 loading ...")
# result = []
# tox_21_list = ['NR-AR', 'NR-AR-LBD', 'NR-AhR', 'NR-Aromatase', 'NR-ER', 'NR-ER-LBD', 'NR-PPAR-gamma', 'SR-ARE', 'SR-ATAD5', 'SR-HSE', 'SR-MMP', 'SR-p53']
# for i, cx in enumerate(tox_21_list):
# result.append(joblib.load('model_saved/Tox21/'+cx+'_rf_model_descriptor.joblib'))
st.markdown(''':rainbow[TOXICITY]''')
# st.write("TOXICITY")
st.write("AMES : ", "Removed due to Github LFS policy")# round(AMES_MACCS.predict(MACCS_fingerprint_1)[0],4))
st.write("Carcinogens : ", "Removed due to Github LFS policy")# round(Carcinogens_Lagunin.predict(MACCS_fingerprint_1)[0],4))
st.write("ClinTox : ", "Removed due to Github LFS policy")# round(ClinTox_MACCS.predict(MACCS_fingerprint_1)[0],4))
st.write("DILI : ", "Removed due to Github LFS policy")# round(DILI_MACCS.predict(MACCS_fingerprint_1)[0],4))
st.write("hERG : ", "Removed due to Github LFS policy")# round(hERG_tox.predict(descriptors_1)[0],4))
st.write("LD50 : ", "Removed due to Github LFS policy")# round(LD50.predict(descriptors_1)[0]),4)
st.write("Skin_Reaction : ", "Removed due to Github LFS policy")# round(Skin_Reaction.predict(MACCS_fingerprint_1)[0],4))
st.write("\n\nTox 21")
# descriptors_without_IPC = descriptors_without_IPC.reset_index()
# descriptors_without_IPC['index'] = descriptors_without_IPC.pop('index')
tox_21_list = ['NR-AR', 'NR-AR-LBD', 'NR-AhR', 'NR-Aromatase', 'NR-ER', 'NR-ER-LBD', 'NR-PPAR-gamma', 'SR-ARE', 'SR-ATAD5', 'SR-HSE', 'SR-MMP', 'SR-p53']
for i, tox_21 in enumerate(tox_21_list):
st.write("Tox 21 with value is :", "Removed due to Github LFS policy")#round(result[i].predict(descriptors_without_IPC2)[0],4))