-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathunittest_atac_seq_analysis.py
161 lines (130 loc) · 6.08 KB
/
unittest_atac_seq_analysis.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
#!/usr/bin/env python
# import modules
import unittest
import os
from pandas.testing import assert_frame_equal
from numpy.testing import assert_array_almost_equal
import ohsu_atac as oa
import pandas as pd
import numpy as np
from sklearn.svm import SVC
from ohsu_atac import process_metadata
from ohsu_atac import normalize_matrix
from ohsu_atac import construct_svc
# define globals
TEST_DIR = os.path.dirname('test/')
TEST_METADATA_FILE = os.path.join(TEST_DIR,'test_metadata.txt')
TEST_ATACSEQ_FILE = os.path.join(TEST_DIR,'test_atac.csv')
# build unit tests
class TestEssentials(unittest.TestCase):
def setUp(self) -> None:
return super().setUp()
def tearDown(self) -> None:
return super().tearDown()
def test_read_metadata(self):
metadata = oa.read_metadata(TEST_METADATA_FILE)
metadata.columns = ['type']
metadata_compare = pd.DataFrame({
'type': ['excitatory_neuron','excitatory_neuron','inhibitory_neuron','excitatory_neuron','inhibitory_neuron']
})
assert_frame_equal(metadata,metadata_compare)
def test_process_metadata(self):
metadata = oa.read_metadata(TEST_METADATA_FILE)
df = pd.read_csv(TEST_ATACSEQ_FILE,index_col=0)
metadata = oa.process_metadata(df,metadata)
metadata_compare = pd.DataFrame(
{
'sample_id':['excitatory_neuron','excitatory_neuron','inhibitory_neuron','excitatory_neuron','inhibitory_neuron'],
'count':[3, 2, 5, 10, 3],
'std':[0.957427,1.000000,2.500000,3.785939,1.500000],
'mean':[0.75,0.50,1.25,2.50,0.75],
'zero':[2,3,3,2,3]
},
index=['GATTCGGTAGTTACGCAAGTCCAA','TGCGGCCTGATCATGAAGCTCGCT','GATTCGGTACCGGAAGCGTTAGAA','TGCGGCCTGCCGGAGCAGTTCAGG','ACGCGACGAATGATGCGATCTATC']
)
assert_frame_equal(metadata,metadata_compare)
def test_filter_matrix(self):
# define filter and comparisons
FILTER_THRESHOLD = 4
FILTERED_DF_COMPARE = pd.DataFrame(
{
'GATTCGGTACCGGAAGCGTTAGAA':[5, 0, 0, 0],
'TGCGGCCTGCCGGAGCAGTTCAGG':[8, 0, 0, 2]
},
index=['chr1-3094454-3095231','chr1-3117883-3118383','chr1-3119738-3120238','chr1-3120562-3121062']
)
FILTERED_METADATA_COMPARE = pd.DataFrame(
{
'sample_id':['inhibitory_neuron','excitatory_neuron'],
'count':[5, 10],
'std':[2.500000,3.785939],
'mean':[1.25,2.50],
'zero':[3, 2]
},
index=['GATTCGGTACCGGAAGCGTTAGAA','TGCGGCCTGCCGGAGCAGTTCAGG']
)
# read in and process parent files
metadata = oa.read_metadata(TEST_METADATA_FILE)
df = pd.read_csv(TEST_ATACSEQ_FILE,index_col=0)
metadata = oa.process_metadata(df,metadata)
# perform filter
filtered_df, filtered_metadata = oa.filter_matrix(df,metadata,filter=FILTER_THRESHOLD)
# confirm equality
assert_frame_equal(filtered_df,FILTERED_DF_COMPARE)
assert_frame_equal(filtered_metadata,FILTERED_METADATA_COMPARE)
def test_normalize_matrix(self):
# define comparison
NORMALIZED_DF_COMPARE = pd.DataFrame({
'GATTCGGTAGTTACGCAAGTCCAA': [2.0,0.0,1.0,0.0],
'TGCGGCCTGATCATGAAGCTCGCT': [3.0,0.0,0.0,0.0],
'GATTCGGTACCGGAAGCGTTAGAA': [3.0,0.0,0.0,0.0],
'TGCGGCCTGCCGGAGCAGTTCAGG': [2.4,0.0,0.0,0.6],
'ACGCGACGAATGATGCGATCTATC': [3.0,0.0,0.0,0.0]
},
index=['chr1-3094454-3095231','chr1-3117883-3118383','chr1-3119738-3120238','chr1-3120562-3121062']
)
# read in parent data and reduce size
df = pd.read_csv(TEST_ATACSEQ_FILE,index_col=0)
# normalize
normalized_df = oa.normalize_matrix(df)
assert_frame_equal(normalized_df,NORMALIZED_DF_COMPARE)
def test_boxcox_transformation(self):
TRANSFORMED_DF_COMPARE = np.array([[ 1.30893484,1.73205081,1.73205081,1.44265337,1.73205081],
[-0.97103679,-0.57735027,-0.57735027,-0.93273282,-0.57735027],
[ 0.63313874,-0.57735027,-0.57735027,-0.93273282,-0.57735027],
[-0.97103679,-0.57735027,-0.57735027,0.42281227,-0.57735027]])
df = pd.read_csv(TEST_ATACSEQ_FILE,index_col=0)
normalized_df = oa.normalize_matrix(df)
transformed_df = oa.boxcox_transformation(normalized_df)
assert_array_almost_equal(transformed_df,TRANSFORMED_DF_COMPARE)
def test_reduce_dimensions(self):
COMPONENTS_COMPARE = np.array([[ 3.57617221, -0.02896694],
[-1.5820112,-0.12770525],
[-0.9695218,1.12654213],
[-1.02463921,-0.96986995]])
df = pd.read_csv(TEST_ATACSEQ_FILE,index_col=0)
normalized_df = oa.normalize_matrix(df)
transformed_df = oa.boxcox_transformation(normalized_df)
components = oa.reduce_dimensions(transformed_df,n_components=2)
assert_array_almost_equal(components,COMPONENTS_COMPARE)
def test_generate_umap(self):
UMAP_COMPONENTS_COMPARE = np.array([[-73.724945,44.733227],
[-72.16196,44.814884],
[-72.917305,44.180145],
[-72.82937,45.469257]]).astype(np.float32)
df = pd.read_csv(TEST_ATACSEQ_FILE,index_col=0)
normalized_df = oa.normalize_matrix(df)
transformed_df = oa.boxcox_transformation(normalized_df)
components = oa.reduce_dimensions(transformed_df,n_components=2)
umap_components = oa.generate_umap(components,random_state=0,n_epochs=10)
assert_array_almost_equal(np.round(umap_components,2),np.round(UMAP_COMPONENTS_COMPARE,2))
def test_construct_svc(self):
pass
df = pd.read_csv(TEST_ATACSEQ_FILE,index_col=0)
normalized_df = oa.normalize_matrix(df)
transformed_df = oa.boxcox_transformation(normalized_df)
components = oa.reduce_dimensions(transformed_df,n_components=2)
umap_components = oa.generate_umap(components,random_state=0,n_epochs=10)
svc = oa.construct_svc()
if __name__ == '__main__':
unittest.main()