-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathdata_loading.py
268 lines (201 loc) · 14.6 KB
/
data_loading.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
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
from csv import reader
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import os
import pickle
import joblib
import argparse
def split_train_val_test(sample_list, ratio = [0.8, 0.1]):
sample_num = len(sample_list)
np.random.shuffle(sample_list)
train_size = int(sample_num * 0.8)
val_size = int(sample_num * 0.1)
test_size = sample_num - train_size - val_size
train_list = sample_list[0:train_size]
val_list = sample_list[train_size+1:train_size+val_size]
test_list = sample_list[train_size+val_size+1:]
return train_list, val_list, test_list
def split_train_test(sample_list, ratio=0.8):
sample_num = len(sample_list)
np.random.shuffle(sample_list)
train_size = int(sample_num * 0.8)
test_size = sample_num - train_size
train_list = sample_list[0:train_size]
test_list = sample_list[train_size:]
return train_list, test_list
def one_hot(a, num_classes):
return np.squeeze(np.eye(num_classes)[a.reshape(-1)])
if __name__ == "__main__":
# specifying data paths
parser = argparse.ArgumentParser(description = 'Data Loading Pipeline',formatter_class=argparse.ArgumentDefaultsHelpFormatter)
parser.add_argument('--data_dir', default='data_stranded', type=str, help='Specify the data directory')
parser.add_argument('--type', default='gene', type=str, help='Specify the hidden nodes of 1st layer')
print("Loading Data")
args = parser.parse_args()
if args.type == 'exon':
data_dir = '../' + args.data_dir
#COPDGene_Freeze1_RNAseq_genes = os.path.join(data_dir, 'COPDGene_Freeze3_RNAseq_genes.csv')
#COPDGene_Freeze1_RNAseq_exonicParts_logCPM_covAdjusted = os.path.join(data_dir, 'COPDGene_Freeze3_RNAseq_exonicParts_logCPM_covAdjustedWOsv.csv')
COPDGene_Freeze1_RNAseq_exonicParts_logCPM_normalized = os.path.join(data_dir, 'COPDGene_Freeze3_RNAseq_exonicParts_logCPM_normalized.csv')
COPDGene_Freeze1_RNAseq_samples = os.path.join(data_dir, 'COPDGene_Freeze3_RNAseq_samples.csv')
# loading csv data to pandas frame
# genes = pd.read_csv(COPDGene_Freeze1_RNAseq_genes)
#exonicParts_logCPM_covAdjusted = pd.read_csv(COPDGene_Freeze1_RNAseq_exonicParts_logCPM_covAdjusted)
print("Reading csv files")
exonicParts_logCPM_normalized = pd.read_csv(COPDGene_Freeze1_RNAseq_exonicParts_logCPM_normalized)
samples = pd.read_csv(COPDGene_Freeze1_RNAseq_samples)
# dealing with data
# so every row represents a patient for now
# shape (520, 266228)
#exonicParts_logCPM_covAdjusted_matrix = np.transpose(exonicParts_logCPM_covAdjusted.values[:, 1:])
print("Loading to matrix")
exonicParts_logCPM_normalized_matrix = np.transpose(exonicParts_logCPM_normalized.values[:, 1:])
# get smoking status
# shape (520, )
smoking_status = samples['SmokCigNow'].values
# TODO:!!!!!
# smoking_status_matrix = one_hot(smoking_status, 2)
# total number of examples
sample_num_total = exonicParts_logCPM_normalized_matrix.shape[0]
assert(smoking_status.shape[0] == sample_num_total)
# get train, val, test indices
pos_list = np.where(smoking_status==1)
neg_list = np.where(smoking_status==0)
train_list_pos, val_list_pos, test_list_pos = split_train_val_test(np.arange(sample_num_total)[pos_list])
train_list_neg, val_list_neg, test_list_neg = split_train_val_test(np.arange(sample_num_total)[neg_list])
#train_list_pos, test_list_pos = split_train_test(np.arange(sample_num_total)[pos_list])
#train_list_neg, test_list_neg = split_train_test(np.arange(sample_num_total)[neg_list])
train_list = np.concatenate((train_list_pos, train_list_neg), axis=0)
val_list = np.concatenate((val_list_pos, val_list_neg), axis=0)
# we keep test as a held-out dataset
test_list = np.concatenate((test_list_pos, test_list_neg), axis=0)
# saving them to proper place using pickle
print("Saving to pickle")
saving_dir = '../' + args.data_dir
#covAdj_path = os.path.join(saving_dir, 'COPDGene_Freeze3_RNAseq_exonicParts_logCPM_covAdjusted')
normalized_path = os.path.join(saving_dir, 'COPDGene_Freeze3_RNAseq_exonicParts_logCPM_normalized')
smoking_path = os.path.join(saving_dir, 'COPDGene_Freeze3_RNAseq_samples', 'SmokCigNow')
#if not os.path.exists(covAdj_path):
# os.mkdir(covAdj_path)
if not os.path.exists(normalized_path):
os.mkdir(normalized_path)
if not os.path.exists(smoking_path):
os.mkdir(smoking_path)
# here we should dump the partition list as well
pickle.dump(train_list, open(os.path.join(saving_dir, 'train_list.pickle'), 'wb'))
pickle.dump(val_list, open(os.path.join(saving_dir, 'val_list.pickle'), 'wb'))
pickle.dump(test_list, open(os.path.join(saving_dir, 'test_list.pickle'), 'wb'))
#pickle.dump(exonicParts_logCPM_covAdjusted_matrix[train_list], open(os.path.join(covAdj_path, 'X_train.pickle'),'wb'))
#pickle.dump(exonicParts_logCPM_covAdjusted_matrix[val_list], open(os.path.join(covAdj_path, 'X_val.pickle'),'wb'))
#pickle.dump(exonicParts_logCPM_covAdjusted_matrix[test_list], open(os.path.join(covAdj_path, 'X_test.pickle'),'wb'))
joblib.dump(exonicParts_logCPM_normalized_matrix[train_list], open(os.path.join(normalized_path, 'X_train.pickle'),'wb'), protocol=2)
joblib.dump(exonicParts_logCPM_normalized_matrix[val_list], open(os.path.join(normalized_path, 'X_val.pickle'),'wb'), protocol=2)
joblib.dump(exonicParts_logCPM_normalized_matrix[test_list], open(os.path.join(normalized_path, 'X_test.pickle'),'wb'), protocol=2)
joblib.dump(smoking_status[train_list], open(os.path.join(smoking_path, 'Y_train.pickle'),'wb'))
joblib.dump(smoking_status[val_list], open(os.path.join(smoking_path, 'Y_val.pickle'),'wb'))
joblib.dump(smoking_status[test_list], open(os.path.join(smoking_path, 'Y_test.pickle'),'wb'))
# here we directly dump X and Y all
#print("Saving all the data for now")
#pickle.dump(exonicParts_logCPM_normalized_matrix, open(os.path.join(normalized_path, 'X_all.pickle'),'wb'))
#pickle.dump(smoking_status, open(os.path.join(smoking_path, 'Y_all.pickle'),'wb'))
elif args.type == 'gene':
data_dir = '../' + args.data_dir
saving_dir = '../' + args.data_dir
train_list = pickle.load(open(os.path.join(saving_dir, 'train_list.pickle'), 'rb'))
val_list = pickle.load(open(os.path.join(saving_dir, 'val_list.pickle'), 'rb'))
test_list = pickle.load(open(os.path.join(saving_dir, 'test_list.pickle'), 'rb'))
#COPDGene_Freeze1_RNAseq_exonicParts_logCPM_covAdjusted = os.path.join(data_dir, 'COPDGene_Freeze3_RNAseq_genes_logCPM_covAdjusted.csv')
COPDGene_Freeze1_RNAseq_exonicParts_logCPM_normalized = os.path.join(data_dir, 'COPDGene_Freeze3_RNAseq_genes_logCPM_normalized.csv')
# loading csv data to pandas frame
#exonicParts_logCPM_covAdjusted = pd.read_csv(COPDGene_Freeze1_RNAseq_exonicParts_logCPM_covAdjusted)
exonicParts_logCPM_normalized = pd.read_csv(COPDGene_Freeze1_RNAseq_exonicParts_logCPM_normalized)
# dealing with data
# so every row represents a patient for now
# shape (520, 266228)
#exonicParts_logCPM_covAdjusted_matrix = np.transpose(exonicParts_logCPM_covAdjusted.values[:, 1:])
exonicParts_logCPM_normalized_matrix = np.transpose(exonicParts_logCPM_normalized.values[:, 1:])
# get smoking status
# shape (520, )
# smoking_status = samples['SmokCigNow'].values
# TODO:!!!!!
# smoking_status_matrix = one_hot(smoking_status, 2)
# total number of examples
sample_num_total = exonicParts_logCPM_normalized_matrix.shape[0]
# assert(smoking_status.shape[0] == sample_num_total)
# get train, val, test indices
#pos_list = np.where(smoking_status==1)
#neg_list = np.where(smoking_status==0)
#train_list_pos, val_list_pos, test_list_pos = split_train_val_test(np.arange(sample_num_total)[pos_list])
#train_list_neg, val_list_neg, test_list_neg = split_train_val_test(np.arange(sample_num_total)[neg_list])
#train_list = np.concatenate((train_list_pos, train_list_neg), axis=0)
#val_list = np.concatenate((val_list_pos, val_list_neg), axis=0)
#test_list = np.concatenate((test_list_pos, test_list_neg), axis=0)
# saving them to proper place using pickle
saving_dir = '../' + args.data_dir
#covAdj_path = os.path.join(saving_dir, 'COPDGene_Freeze1_RNAseq_exonicParts_logCPM_covAdjusted')
normalized_path = os.path.join(saving_dir, 'COPDGene_Freeze3_RNAseq_genes_logCPM_normalized')
#smoking_path = os.path.join(saving_dir, 'COPDGene_Freeze1_RNAseq_samples', 'SmokCigNow')
#pickle.dump(exonicParts_logCPM_covAdjusted_matrix[train_list], open(os.path.join(covAdj_path, 'X_train.pickle'),'wb'))
#pickle.dump(exonicParts_logCPM_covAdjusted_matrix[val_list], open(os.path.join(covAdj_path, 'X_val.pickle'),'wb'))
#pickle.dump(exonicParts_logCPM_covAdjusted_matrix[test_list], open(os.path.join(covAdj_path, 'X_test.pickle'),'wb'))
joblib.dump(exonicParts_logCPM_normalized_matrix[train_list], open(os.path.join(normalized_path, 'X_train.pickle'),'wb'), protocol=2)
joblib.dump(exonicParts_logCPM_normalized_matrix[val_list], open(os.path.join(normalized_path, 'X_val.pickle'),'wb'), protocol=2)
joblib.dump(exonicParts_logCPM_normalized_matrix[test_list], open(os.path.join(normalized_path, 'X_test.pickle'),'wb'), protocol=2)
#pickle.dump(smoking_status[train_list], open(os.path.join(smoking_path, 'Y_train.pickle'),'wb'))
#pickle.dump(smoking_status[val_list], open(os.path.join(smoking_path, 'Y_val.pickle'),'wb'))
#pickle.dump(smoking_status[test_list], open(os.path.join(smoking_path, 'Y_test.pickle'),'wb'))
# here we directly dump X and Y all
#print("Saving all the data for now")
#pickle.dump(exonicParts_logCPM_normalized_matrix, open(os.path.join(normalized_path, 'X_all.pickle'),'wb'))
#pickle.dump(smoking_status, open(os.path.join(smoking_path, 'Y_all.pickle'),'wb'))
elif args.type == 'trans':
data_dir = '../' + args.data_dir
saving_dir = '../' + args.data_dir
train_list = pickle.load(open(os.path.join(saving_dir, 'train_list.pickle'), 'rb'))
val_list = pickle.load(open(os.path.join(saving_dir, 'val_list.pickle'), 'rb'))
test_list = pickle.load(open(os.path.join(saving_dir, 'test_list.pickle'), 'rb'))
COPDGene_Freeze1_RNAseq_exonicParts_logCPM_normalized = os.path.join(data_dir, 'COPDGene_Freeze3_RNAseq_transcripts_logCPM_normalized.csv')
# loading csv data to pandas frame
#exonicParts_logCPM_covAdjusted = pd.read_csv(COPDGene_Freeze1_RNAseq_exonicParts_logCPM_covAdjusted)
exonicParts_logCPM_normalized = pd.read_csv(COPDGene_Freeze1_RNAseq_exonicParts_logCPM_normalized)
# dealing with data
# so every row represents a patient for now
# shape (520, 266228)
# exonicParts_logCPM_covAdjusted_matrix = np.transpose(exonicParts_logCPM_covAdjusted.values[:, 1:])
exonicParts_logCPM_normalized_matrix = np.transpose(exonicParts_logCPM_normalized.values[:, 1:])
# get smoking status
# shape (520, )
# smoking_status = samples['SmokCigNow'].values
# TODO:!!!!!
# smoking_status_matrix = one_hot(smoking_status, 2)
# total number of examples
# sample_num_total = exonicParts_logCPM_covAdjusted_matrix.shape[0]
# assert(smoking_status.shape[0] == sample_num_total)
# get train, val, test indices
#pos_list = np.where(smoking_status==1)
#neg_list = np.where(smoking_status==0)
#train_list_pos, val_list_pos, test_list_pos = split_train_val_test(np.arange(sample_num_total)[pos_list])
#train_list_neg, val_list_neg, test_list_neg = split_train_val_test(np.arange(sample_num_total)[neg_list])
#train_list = np.concatenate((train_list_pos, train_list_neg), axis=0)
#val_list = np.concatenate((val_list_pos, val_list_neg), axis=0)
#test_list = np.concatenate((test_list_pos, test_list_neg), axis=0)
# saving them to proper place using pickle
saving_dir = '../' + args.data_dir
#covAdj_path = os.path.join(saving_dir, 'COPDGene_Freeze1_RNAseq_exonicParts_logCPM_covAdjusted')
normalized_path = os.path.join(saving_dir, 'COPDGene_Freeze3_RNAseq_transcripts_logCPM_normalized')
joblib.dump(exonicParts_logCPM_normalized_matrix[train_list], open(os.path.join(normalized_path, 'X_train.pickle'),'wb'), protocol=2)
joblib.dump(exonicParts_logCPM_normalized_matrix[val_list], open(os.path.join(normalized_path, 'X_val.pickle'),'wb'), protocol=2)
joblib.dump(exonicParts_logCPM_normalized_matrix[test_list], open(os.path.join(normalized_path, 'X_test.pickle'),'wb'), protocol=2)
#smoking_path = os.path.join(saving_dir, 'COPDGene_Freeze1_RNAseq_samples', 'SmokCigNow')
#pickle.dump(exonicParts_logCPM_covAdjusted_matrix[train_list], open(os.path.join(covAdj_path, 'X_train.pickle'),'wb'))
#pickle.dump(exonicParts_logCPM_covAdjusted_matrix[val_list], open(os.path.join(covAdj_path, 'X_val.pickle'),'wb'))
#pickle.dump(exonicParts_logCPM_covAdjusted_matrix[test_list], open(os.path.join(covAdj_path, 'X_test.pickle'),'wb'))
#pickle.dump(exonicParts_logCPM_normalized_matrix[train_list], open(os.path.join(normalized_path, 'X_train.pickle'),'wb'))
#pickle.dump(exonicParts_logCPM_normalized_matrix[val_list], open(os.path.join(normalized_path, 'X_val.pickle'),'wb'))
#pickle.dump(exonicParts_logCPM_normalized_matrix[test_list], open(os.path.join(normalized_path, 'X_test.pickle'),'wb'))
#pickle.dump(smoking_status[train_list], open(os.path.join(smoking_path, 'Y_train.pickle'),'wb'))
#pickle.dump(smoking_status[val_list], open(os.path.join(smoking_path, 'Y_val.pickle'),'wb'))
#pickle.dump(smoking_status[test_list], open(os.path.join(smoking_path, 'Y_test.pickle'),'wb'))
# here we directly dump X and Y all
#pickle.dump(smoking_status, open(os.path.join(smoking_path, 'Y_all.pickle'),'wb'))