-
Notifications
You must be signed in to change notification settings - Fork 0
/
Test02_BeatAML.py
356 lines (265 loc) · 15.2 KB
/
Test02_BeatAML.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
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
#!/usr/bin/env python
# coding: utf-8
# In[ ]:
import os
# import sys
import numpy as np
# import scipy as sp
import pandas as pd
# import scipy.linalg as alg
import pickle
# import matplotlib.pyplot as plt
from time import time
from tqdm import tqdm
from skrebate import ReliefF
from sklearn.model_selection import KFold
from sklearn.svm import SVR as SupportVectorRegressor
from sklearn.linear_model import ElasticNet
from sklearn.ensemble import RandomForestRegressor
from scipy.stats import pearsonr, spearmanr
## Path & Filename...
PATH = os.path.join(os.getenv("HOMEPATH"), "Google Drive\\Study\\ECE 5332-009 - Topics in EE, Data Science\\BeatAML\\")
DIRNAME = ("Training\\", "Leaderboard\\")
FILENAME = ("rnaseq.csv", "dnaseq.csv", "clinical_numerical.csv", "clinical_categorical.csv",
"clinical_categorical_legend.csv", "aucs.csv", "response.csv")
os.chdir(PATH)
### Read data...
## Training...
RNA = pd.read_csv(PATH + DIRNAME[0] + FILENAME[0], header = 0)
# DNA = pd.read_csv(PATH + DIRNAME[0] + FILENAME[1], header = 0)
# CLI_NUM = pd.read_csv(PATH + DIRNAME[0] + FILENAME[2], header = 0)
# CLI_CAT = pd.read_csv(PATH + DIRNAME[0] + FILENAME[3], header = 0)
# CLI_CAT_LEG = pd.read_csv(PATH + DIRNAME[0] + FILENAME[4], header = 0)
AUC = pd.read_csv(PATH + DIRNAME[0] + FILENAME[5], header = 0)
# RESP = pd.read_csv(PATH + DIRNAME[0] + FILENAME[6], header = 0)
## Leaderboard...
RNA_LB = pd.read_csv(PATH + DIRNAME[1] + FILENAME[0], header = 0)
# DNA_LB = pd.read_csv(PATH + DIRNAME[1] + FILENAME[1], header = 0)
# CLI_NUM_LB = pd.read_csv(PATH + DIRNAME[1] + FILENAME[2], header = 0)
# CLI_CAT_LB = pd.read_csv(PATH + DIRNAME[1] + FILENAME[3], header = 0)
# CLI_CAT_LEG_LB = pd.read_csv(PATH + DIRNAME[1] + FILENAME[4], header = 0)
AUC_LB = pd.read_csv(PATH + DIRNAME[1] + FILENAME[5], header = 0)
# RESP_LB = pd.read_csv(PATH + DIRNAME[1] + FILENAME[6], header = 0)
## Processing...
all(AUC.inhibitor.unique() == AUC_LB.inhibitor.unique()) ## Check if same drugs
all(RNA[["Gene", "Symbol"]] == RNA_LB[["Gene", "Symbol"]]) ## Check if same genes
lab_id_list = dict(TR = RNA.columns[2:], LB = RNA_LB.columns[2:])
gene_list, drug_list = RNA[["Gene", "Symbol"]], AUC.inhibitor.unique().tolist()
RNA.index, RNA_LB.index = gene_list.Symbol, gene_list.Symbol
RNA, RNA_LB = RNA.iloc[:, 2:], RNA_LB.iloc[:, 2:]
# RNA.shape, RNA_LB.shape
var_idx = (RNA.var(axis = 1) > 0.1).to_numpy() ## Filter genes w/ low variability
RNA_filt, RNA_LB_filt = RNA.iloc[var_idx, :], RNA_LB.iloc[var_idx, :]
# In[ ]:
## Function definitions...
def RF(X_train, y_train, X_test, seed = 0):
mdl = RandomForestRegressor(n_estimators = 200, criterion = "mae", min_samples_leaf = 5, random_state = seed)
y_pred = mdl.fit(X_train, y_train).predict(X_test)
return y_pred
def SVR(X_train, y_train, X_test):
mdl = SupportVectorRegressor(kernel = "poly", degree = 3, gamma = "scale", tol = 1e-3, C = 100)
y_pred = mdl.fit(X_train, y_train).predict(X_test)
return y_pred
def EN(X_train, y_train, X_test, seed):
mdl = ElasticNet(fit_intercept = True, l1_ratio = 0.8, alpha = 0.2, tol = 1e-3, random_state = seed)
y_pred = mdl.fit(X_train, y_train).predict(X_test)
return y_pred
def EVAL_PERF(y_label, y_pred, alpha = 0.05):
y_label, y_pred = np.array(y_label).squeeze(), np.array(y_pred).squeeze()
PCC, pval = pearsonr(y_label, y_pred); #PCC = PCC if pval < alpha else 0
SCC, pval = spearmanr(y_label, y_pred); #SCC = SCC if pval < alpha else 0
NRMSE = np.sqrt(((y_label - y_pred)**2).mean()) / y_label.std(ddof = 0)
NMAE = (np.abs(y_label - y_pred)).mean() / (np.abs(y_label - y_label.mean())).mean()
return PCC, SCC, NRMSE, NMAE
# In[ ]:
### ORIGINAL ANALYSIS CELL...
drug_list = AUC.inhibitor.unique().tolist()
RNA_filt = RNA.iloc[(RNA.var(axis = 1) > 0.1).to_numpy(), :]
FS = ReliefF(n_features_to_select = 1000, n_neighbors = 10, n_jobs = 1); p_top = 1500
feature_ranks = {kk: [ ] for kk in drug_list}
RESULTS_RF = pd.DataFrame(dtype = float, index = drug_list, columns = ["PCC", "SCC", "NRMSE", "NMAE"])
RESULTS_SVR = pd.DataFrame(dtype = float, index = drug_list, columns = ["PCC", "SCC", "NRMSE", "NMAE"])
RESULTS_EN = pd.DataFrame(dtype = float, index = drug_list, columns = ["PCC", "SCC", "NRMSE", "NMAE"])
RESULTS_ENS = pd.DataFrame(dtype = float, index = drug_list, columns = ["PCC", "SCC", "NRMSE", "NMAE"])
N = len(drug_list); count = 76
for drug in tqdm(drug_list[76:N]): ## Continue from where left off
count +=1; print("\nChosen drug# = %d: %s" % (count, drug))
y_data = AUC.iloc[(AUC.inhibitor == drug).tolist(), :]
X_data, y_data = RNA_filt.loc[:, y_data.lab_id].T, y_data["auc"].to_numpy()
Y_pred = pd.DataFrame(dtype = float, index = X_data.index, columns = ["Actual", "RF", "SVR", "EN", "ENS"])
CV = KFold(n_splits = 3, shuffle = False, random_state = 0)
dt = time()
for train_idx, test_idx in CV.split(X_data):
X_train, y_train = X_data.iloc[train_idx, :], y_data[train_idx]
X_test, y_test = X_data.iloc[test_idx, :], y_data[test_idx]
## Perform ReliefF...
# dt = time();
FS.fit(X_train.values, y_train)
# dt = time() - dt; print("Elapsed time = %0.4f sec." % dt)
feat_top = FS.top_features_; feature_ranks[drug].append(feat_top)
X_train, X_test = X_train.iloc[:, feat_top[:p_top]], X_test.iloc[:, feat_top[:p_top]]
## Perform prediction...
test_idx_lab = X_data.index[test_idx]
Y_pred.loc[test_idx_lab, "Actual"] = y_test
Y_pred.loc[test_idx_lab, "RF"] = RF(X_train, y_train, X_test, seed = 0)
Y_pred.loc[test_idx_lab, "SVR"] = SVR(X_train, y_train, X_test)
Y_pred.loc[test_idx_lab, "EN"] = EN(X_train, y_train, X_test, seed = 0)
# Y_pred.loc[test_idx_lab, "ENS"] = Y_pred.loc[test_idx_lab, ["RF", "SVR", "EN"]].mean(axis = 1)
#### CV loop ends.
dt = time() - dt; print("Elapsed time = %0.4f sec." % dt)
## Ensemble prediction...
Y_pred.loc[:, "ENS"] = Y_pred.loc[:, ["RF", "SVR", "EN"]].mean(axis = 1)
## Save results...
# PCC, SCC, NEMSE, NMAE = EVAL_PERF(Y_pred["Actual"], Y_pred["RF"])
RESULTS_RF.loc[drug, :] = EVAL_PERF(Y_pred["Actual"], Y_pred["RF"])
RESULTS_SVR.loc[drug, :] = EVAL_PERF(Y_pred["Actual"], Y_pred["SVR"])
RESULTS_EN.loc[drug, :] = EVAL_PERF(Y_pred["Actual"], Y_pred["EN"])
RESULTS_ENS.loc[drug, :] = EVAL_PERF(Y_pred["Actual"], Y_pred["ENS"])
## Display results...
RES_TAB = pd.concat((RESULTS_RF.loc[drug, :], RESULTS_SVR.loc[drug, :], RESULTS_EN.loc[drug, :], RESULTS_ENS.loc[drug, :]),
axis = 1, ignore_index = True)
RES_TAB.columns = ["RF", "SVR", "EN", "ENS"]; print("\n", RES_TAB)
# print(RESULTS_RF.loc[drug, :])
# print(RESULTS_SVR.loc[drug, :])
# print(RESULTS_EN.loc[drug, :])
####
# In[ ]:
# %qtconsole --style monokai
with open("FS1_Drugs_25.pickle", "wb") as file:
pickle.dump(feature_ranks, file, protocol = pickle.HIGHEST_PROTOCOL)
with open("FS2_Drugs_50.pickle", "wb") as file:
pickle.dump(feature_ranks, file, protocol = pickle.HIGHEST_PROTOCOL)
with open("FS1_Drugs_25_gn.pickle", "rb") as file:
aa1 = pickle.load(file)
with open("FS2_Drugs_51.pickle", "rb") as file:
aa2 = pickle.load(file)
aa3 = {kk: [ ] for kk in aa1.keys()}
for kk in aa2.keys():
if len(aa1[kk]) == 3:
aa3[kk] = aa1[kk]
elif len(aa2[kk]) == 3:
aa3[kk] = aa2[kk]
aa3["gene_names"] = aa1["gene_names"]
with open("FS_Drugs_76.pickle", "wb") as file:
pickle.dump(aa3, file, protocol = pickle.HIGHEST_PROTOCOL)
with open("FS_Drugs_76.pickle", "rb") as file:
feature_ranks = pickle.load(file)
with open("FS_Drugs_122_ReliefF_3_folds.pickle", "wb") as file:
pickle.dump(feature_ranks, file, protocol = pickle.HIGHEST_PROTOCOL)
# In[ ]:
### CELL FOR AFTER GETTING FEATURE RANKS...
# drug_list = AUC.inhibitor.unique().tolist()
# RNA_filt = RNA.iloc[(RNA.var(axis = 1) > 0.1).to_numpy(), :]
# FS = ReliefF(n_features_to_select = 1000, n_neighbors = 10, n_jobs = 1);
p_top = 100
# feature_ranks = {kk: [ ] for kk in drug_list}
# with open("FS_Drugs_76.pickle", "rb") as file:
# feature_ranks = pickle.load(file)
with open("FS_Drugs_122_ReliefF_3_folds.pickle", "rb") as file:
feature_ranks = pickle.load(file)
RESULTS_RF = pd.DataFrame(dtype = float, index = drug_list, columns = ["PCC", "SCC", "NRMSE", "NMAE"])
RESULTS_SVR = pd.DataFrame(dtype = float, index = drug_list, columns = ["PCC", "SCC", "NRMSE", "NMAE"])
RESULTS_EN = pd.DataFrame(dtype = float, index = drug_list, columns = ["PCC", "SCC", "NRMSE", "NMAE"])
RESULTS_ENS = pd.DataFrame(dtype = float, index = drug_list, columns = ["PCC", "SCC", "NRMSE", "NMAE"])
N = len(drug_list); count = 0
dt = time()
for drug in tqdm(drug_list[:N]): ## Continue from where left off
count += 1; # print("\nChosen drug# = %d: %s" % (count, drug))
y_data = AUC.iloc[(AUC.inhibitor == drug).tolist(), :]
X_data, y_data = RNA_filt.loc[:, y_data.lab_id].T, y_data["auc"].to_numpy()
Y_pred = pd.DataFrame(dtype = float, index = X_data.index, columns = ["Actual", "RF", "SVR", "EN", "ENS"])
CV = KFold(n_splits = 3, shuffle = False, random_state = 0)
fold = 0; # dt = time()
for train_idx, test_idx in CV.split(X_data):
X_train, y_train = X_data.iloc[train_idx, :], y_data[train_idx]
X_test, y_test = X_data.iloc[test_idx, :], y_data[test_idx]
## Perform ReliefF...
# dt = time();
# FS.fit(X_train.values, y_train)
# dt = time() - dt; print("Elapsed time = %0.4f sec." % dt)
# feat_top = FS.top_features_; feature_ranks[drug].append(feat_top)
feat_top = feature_ranks[drug][fold]
X_train, X_test = X_train.iloc[:, feat_top[:p_top]], X_test.iloc[:, feat_top[:p_top]]
## Perform prediction...
test_idx_lab = X_data.index[test_idx]
Y_pred.loc[test_idx_lab, "Actual"] = y_test
Y_pred.loc[test_idx_lab, "RF"] = RF(X_train, y_train, X_test, seed = 0)
Y_pred.loc[test_idx_lab, "SVR"] = SVR(X_train, y_train, X_test)
Y_pred.loc[test_idx_lab, "EN"] = EN(X_train, y_train, X_test, seed = 0)
# Y_pred.loc[test_idx_lab, "ENS"] = Y_pred.loc[test_idx_lab, ["RF", "SVR", "EN"]].mean(axis = 1)
fold += 1
#### CV loop ends.
# dt = time() - dt; print("Elapsed time = %0.4f sec." % dt)
## Ensemble prediction...
Y_pred.loc[:, "ENS"] = Y_pred.loc[:, ["RF", "SVR", "EN"]].mean(axis = 1)
## Save results...
# PCC, SCC, NEMSE, NMAE = EVAL_PERF(Y_pred["Actual"], Y_pred["RF"])
RESULTS_RF.loc[drug, :] = EVAL_PERF(Y_pred["Actual"], Y_pred["RF"])
RESULTS_SVR.loc[drug, :] = EVAL_PERF(Y_pred["Actual"], Y_pred["SVR"])
RESULTS_EN.loc[drug, :] = EVAL_PERF(Y_pred["Actual"], Y_pred["EN"])
RESULTS_ENS.loc[drug, :] = EVAL_PERF(Y_pred["Actual"], Y_pred["ENS"])
# ## Display results...
# RES_TAB = pd.concat((RESULTS_RF.loc[drug, :], RESULTS_SVR.loc[drug, :], RESULTS_EN.loc[drug, :],
# RESULTS_ENS.loc[drug, :]), axis = 1, ignore_index = True)
# RES_TAB.columns = ["RF", "SVR", "EN", "ENS"]; print("\n", RES_TAB)
# print(RESULTS_RF.loc[drug, :])
# print(RESULTS_SVR.loc[drug, :])
# print(RESULTS_EN.loc[drug, :])
#### Whole loop ends...
dt = time() - dt; print("Elapsed time = %0.4f sec." % dt)
RESULTS_MEAN = pd.concat(((RESULTS_RF.iloc[:N, :]).mean(axis = 0), (RESULTS_SVR.iloc[:N, :]).mean(axis = 0),
(RESULTS_EN.iloc[:N, :]).mean(axis = 0), (RESULTS_ENS.iloc[:N, :]).mean(axis = 0)),
axis = 1, ignore_index = True)
RESULTS_MEAN.columns = ["RF", "SVR", "EN", "ENS"]; print("Mean performance for %d inhibitors = \n" % N, RESULTS_MEAN)
# In[ ]
### TEST ON LEADERBOARD DATA...
p_top = 500
# feature_ranks = {kk: [ ] for kk in drug_list}
with open("FS_Drugs_122_ReliefF_3_folds.pickle", "rb") as file:
feature_ranks = pickle.load(file)
RESULTS_LB_RF = pd.DataFrame(dtype = float, index = drug_list, columns = ["PCC", "SCC", "NRMSE", "NMAE"])
RESULTS_LB_SVR = pd.DataFrame(dtype = float, index = drug_list, columns = ["PCC", "SCC", "NRMSE", "NMAE"])
RESULTS_LB_EN = pd.DataFrame(dtype = float, index = drug_list, columns = ["PCC", "SCC", "NRMSE", "NMAE"])
RESULTS_LB_ENS = pd.DataFrame(dtype = float, index = drug_list, columns = ["PCC", "SCC", "NRMSE", "NMAE"])
N = len(drug_list); count = 0
dt = time()
for drug in tqdm(drug_list[:N]): ## Continue from where left off
count += 1; # print("\nChosen drug# = %d: %s" % (count, drug))
y_TR, y_LB = AUC.iloc[(AUC.inhibitor == drug).tolist(), :], AUC_LB.iloc[(AUC_LB.inhibitor == drug).tolist(), :]
X_TR, y_TR = RNA_filt.loc[:, y_TR.lab_id].T, y_TR["auc"].to_numpy()
X_LB, y_LB = RNA_LB_filt.loc[:, y_LB.lab_id].T, y_LB["auc"].to_numpy()
Y_LB_pred = pd.DataFrame(dtype = float, index = X_LB.index, columns = ["Actual", "RF", "SVR", "EN", "ENS"])
fold = 0; # dt = time()
# ## Perform ReliefF...
# dt = time(); FS.fit(X_TR.values, y_TR); dt = time() - dt; print("Elapsed time = %0.4f sec." % dt)
# feat_top = FS.top_features_; feature_ranks[drug].append(feat_top)
## From saved file...
feat_top = feature_ranks[drug][fold]
X_TR, X_LB = X_TR.iloc[:, feat_top[:p_top]], X_LB.iloc[:, feat_top[:p_top]]
## Perform prediction...
Y_LB_pred.loc[:, "Actual"] = y_LB
Y_LB_pred.loc[:, "RF"] = RF(X_TR, y_TR, X_LB, seed = 0)
Y_LB_pred.loc[:, "SVR"] = SVR(X_TR, y_TR, X_LB)
Y_LB_pred.loc[:, "EN"] = EN(X_TR, y_TR, X_LB, seed = 0)
# fold += 1
# dt = time() - dt; print("Elapsed time = %0.4f sec." % dt)
## Ensemble prediction...
Y_LB_pred.loc[:, "ENS"] = Y_LB_pred.loc[:, ["RF", "SVR", "EN"]].mean(axis = 1)
## Save results...
RESULTS_LB_RF.loc[drug, :] = EVAL_PERF(Y_LB_pred["Actual"], Y_LB_pred["RF"])
RESULTS_LB_SVR.loc[drug, :] = EVAL_PERF(Y_LB_pred["Actual"], Y_LB_pred["SVR"])
RESULTS_LB_EN.loc[drug, :] = EVAL_PERF(Y_LB_pred["Actual"], Y_LB_pred["EN"])
RESULTS_LB_ENS.loc[drug, :] = EVAL_PERF(Y_LB_pred["Actual"], Y_LB_pred["ENS"])
# ## Display results...
# RES_LB_TAB = pd.concat((RESULTS_LB_RF.loc[drug, :], RESULTS_LB_SVR.loc[drug, :], RESULTS_LB_EN.loc[drug, :],
# RESULTS_LB_ENS.loc[drug, :]), axis = 1, ignore_index = True)
# RES_LB_TAB.columns = ["RF", "SVR", "EN", "ENS"]; print("\n", RES_LB_TAB)
# print(RESULTS_RF.loc[drug, :])
# print(RESULTS_SVR.loc[drug, :])
# print(RESULTS_EN.loc[drug, :])
#### Whole loop ends...
dt = time() - dt; print("Elapsed time = %0.4f sec." % dt)
RESULTS_LB_MEAN = pd.concat(((RESULTS_LB_RF.iloc[:N, :]).mean(axis = 0), (RESULTS_LB_SVR.iloc[:N, :]).mean(axis = 0),
(RESULTS_LB_EN.iloc[:N, :]).mean(axis = 0), (RESULTS_LB_ENS.iloc[:N, :]).mean(axis = 0)),
axis = 1, ignore_index = True)
RESULTS_LB_MEAN.columns = ["RF", "SVR", "EN", "ENS"]; print("Mean performance for %d inhibitors = \n" % N, RESULTS_LB_MEAN)