-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathmain.py
112 lines (103 loc) · 6.98 KB
/
main.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
#!/usr/bin/python3
import joblib
import reader
import argparse as ap
import model_en, model_lr, model_rf
parser = ap.ArgumentParser()
parser.add_argument('-norm', type=str, required=True, help='<gct file of gene norm>')
parser.add_argument('-att', type=str, required=True, help='<tsv file of sample attributes>')
parser.add_argument('-pheno', type=str, required=True, help='<tsv file of subject phenotypes>')
#parser.add_argument('-srrgtex', type=str, required=True, help='<tsv file of SRR id with corresponding GTEx id>')
parser.add_argument('-ginf', type=str, required=True, help='<tsv file of human gene info>')
parser.add_argument('-sf', type=str, required=True, help='<tsv file of splicing factors>')
parser.add_argument('-psi', type=str, required=True, help='<tsv file of PSI values>')
parser.add_argument('-model', type=str, required=True, help='Model type used for prediction, [\"en\": Elastic Net regression | \"lr\": LASSO regression | \"rf\": Random Forest regression]')
parser.add_argument('-process_data_flag', type=str, required=True, help='Set to [True] to process all data, if set to [False] then simply load the processed data object.')
parser.add_argument('-compute_result_flag', type=str, required=True, help='Set to [True] to train and test model, compute result, if set to [False] then simply load the computed result object.')
parser.add_argument('-dimension_reduct_method', type=str, required=True, help='Type of data dimensionality reduction method, [\"pca\": principal component analysis | \"cluster\": hierarchical clustering]')
args = parser.parse_args()
def main():
# ------------------------------------
# read data from files, merge useful data in a DataFrame
# ------------------------------------
data_obj_file_name_pca = '/nfs/home/students/ge52qoj/SFEEBoT/output/processed_data_pca.sav'
data_obj_file_name_cluster = '/nfs/home/students/ge52qoj/SFEEBoT/output/processed_data_cluster.sav'
if args.process_data_flag.upper() == 'TRUE':
if args.dimension_reduct_method.lower() == 'pca':
dfs = reader.read_all(norm=args.norm, att=args.att, pheno=args.pheno, ginf=args.ginf, sf=args.sf, psi=args.psi, drm=args.dimension_reduct_method.lower()) #srrgtex=args.srrgtex,
tissue_dfs = reader.merge_all(dfs)
wb_sf_dfs = reader.process_tissue_wise(tissue_dfs)
joblib.dump(wb_sf_dfs, data_obj_file_name_pca)
elif args.dimension_reduct_method.lower() == 'cluster':
dfs = reader.read_all(norm=args.norm, att=args.att, pheno=args.pheno, ginf=args.ginf, sf=args.sf, psi=args.psi, drm=args.dimension_reduct_method.lower()) #srrgtex=args.srrgtex,
tissue_dfs = reader.merge_all(dfs)
wb_sf_dfs = reader.process_tissue_wise(tissue_dfs)
joblib.dump(wb_sf_dfs, data_obj_file_name_cluster)
if args.process_data_flag.upper() == 'FALSE':
if args.dimension_reduct_method.lower() == 'pca':
wb_sf_dfs = joblib.load(data_obj_file_name_pca)
elif args.dimension_reduct_method.lower() == 'cluster':
wb_sf_dfs = joblib.load(data_obj_file_name_cluster)
# ------------------------------------
# fit data to model, training & predicting & validating
# ------------------------------------
model = args.model
if model == 'en':
result_obj_file_name_pca = '/nfs/home/students/ge52qoj/SFEEBoT/output/result_en_pca.sav'
result_obj_file_name_cluster = '/nfs/home/students/ge52qoj/SFEEBoT/output/result_en_cluster.sav'
if args.compute_result_flag.upper() == 'TRUE':
if args.dimension_reduct_method.lower() == 'pca':
tissue_2_result_df = model_en.run_single_sf_pca(wb_sf_dfs)
joblib.dump(tissue_2_result_df, result_obj_file_name_pca)
model_en.analyse_result_pca(tissue_2_result_df)
elif args.dimension_reduct_method.lower() == 'cluster':
results = model_en.run_single_sf_cluster(wb_sf_dfs)
joblib.dump(results, result_obj_file_name_cluster)
model_en.analyse_result_cluster(results)
elif args.compute_result_flag.upper() == 'FALSE':
if args.dimension_reduct_method.lower() == 'pca':
tissue_2_result_df = joblib.load(result_obj_file_name_pca)
model_en.analyse_result_pca(tissue_2_result_df)
elif args.dimension_reduct_method.lower() == 'cluster':
results = joblib.load(result_obj_file_name_cluster)
model_en.analyse_result_cluster(results)
if model == 'lr':
result_obj_file_name_pca = '/nfs/home/students/ge52qoj/SFEEBoT/output/result_lr_pca.sav'
result_obj_file_name_cluster = '/nfs/home/students/ge52qoj/SFEEBoT/output/result_lr_cluster.sav'
if args.compute_result_flag.upper() == 'TRUE':
if args.dimension_reduct_method.lower() == 'pca':
tissue_2_result_df = model_lr.run_single_sf_pca(wb_sf_dfs)
joblib.dump(tissue_2_result_df, result_obj_file_name_pca)
model_lr.analyse_result_pca(tissue_2_result_df)
elif args.dimension_reduct_method.lower() == 'cluster':
results = model_lr.run_single_sf_cluster(wb_sf_dfs)
joblib.dump(results, result_obj_file_name_cluster)
model_lr.analyse_result_cluster(results)
elif args.compute_result_flag.upper() == 'FALSE':
if args.dimension_reduct_method.lower() == 'pca':
tissue_2_result_df = joblib.load(result_obj_file_name_pca)
model_lr.analyse_result_pca(tissue_2_result_df)
elif args.dimension_reduct_method.lower() == 'cluster':
results = joblib.load(result_obj_file_name_cluster)
model_lr.analyse_result_cluster(results)
if model == 'rf':
result_obj_file_name_pca = '/nfs/home/students/ge52qoj/SFEEBoT/output/result_rf_pca.sav'
result_obj_file_name_cluster = '/nfs/home/students/ge52qoj/SFEEBoT/output/result_rf_cluster.sav'
if args.compute_result_flag.upper() == 'TRUE':
if args.dimension_reduct_method.lower() == 'pca':
tissue_2_result_df = model_rf.run_single_sf_pca(wb_sf_dfs)
joblib.dump(tissue_2_result_df, result_obj_file_name_pca)
model_rf.analyse_result_pca(tissue_2_result_df)
elif args.dimension_reduct_method.lower() == 'cluster':
results = model_rf.run_single_sf_cluster(wb_sf_dfs)
joblib.dump(results, result_obj_file_name_cluster)
model_rf.analyse_result_cluster(results)
elif args.compute_result_flag.upper() == 'FALSE':
if args.dimension_reduct_method.lower() == 'pca':
tissue_2_result_df = joblib.load(result_obj_file_name_pca)
model_rf.analyse_result_pca(tissue_2_result_df)
elif args.dimension_reduct_method.lower() == 'cluster':
results = joblib.load(result_obj_file_name_cluster)
model_rf.analyse_result_cluster(results)
if __name__ == '__main__':
main()