-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathcv_effect_sizes_continuous.py
175 lines (142 loc) · 6.1 KB
/
cv_effect_sizes_continuous.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
import os
#import util
import numpy as np
import pandas as pd
from argparse import ArgumentParser
import random
import patsy as pat
import statsmodels.api as sm
from sklearn.preprocessing import StandardScaler
from statsmodels.stats.multitest import multipletests
from sklearn.model_selection import StratifiedKFold
def standardize(mask,data):
scaler = StandardScaler(with_mean=False, with_std=True)
scaler.fit(data[mask])
standardized=scaler.transform(data)
return standardized
# redefined function from util to only generate standardized betamaps
def variable_effect(pheno,var,regressors,conn):
"""
Test effect of continuous variable.
pheno = dataframe:
-filtered to be only relevant subjects (use mask_var)
var = column from pheno
regressors = list of strings, formatted for patsy
connectomes = n_subjects x n_edges array
Returns:
table = n_edges
- betas_std = including standardization on controls
- pvalues = pvalues
- qvalues = fdr corrected pvalues alpha = 0.05
"""
n_edges = conn.shape[1]
contrast = np.zeros(1 + len(regressors))
contrast[0] = 1
betas_std = np.zeros(n_edges)
pvalues = np.zeros(n_edges)
formula = ' + '.join((regressors + [var]))
dmat = pat.dmatrix(formula, pheno, return_type='dataframe',NA_action='raise')
mask_std = np.ones(pheno.shape[0]).astype(bool)
conn_std = standardize(mask_std, conn)
for edge in range(n_edges):
model_std = sm.OLS(conn_std[:,edge],dmat)
results_std = model_std.fit()
betas_std[edge] = results_std.params[var]
pvalues[edge] = results_std.pvalues[var]
mt = multipletests(pvalues,method='fdr_bh')
reject = mt[0]
qvalues = mt[1]
table = pd.DataFrame(np.array([betas_std,pvalues,qvalues,reject]).transpose(),
columns=['betas_std','pvalues','qvalues','reject'])
return table
if __name__ == "__main__":
parser = ArgumentParser()
parser.add_argument("--path_pheno",help="path to phenotype .csv file",dest='path_pheno')
parser.add_argument("--path_connectomes",help="path to connectomes .csv file",dest='path_connectomes')
parser.add_argument("--path_out",help="path to output directory",dest='path_out')
parser.add_argument("--n_folds",help="Number of folds of cross_validation",dest='n_folds',type=int,default=5)
args = parser.parse_args()
#############
# LOAD DATA #
#############
path_pheno = args.path_pheno
path_connectomes = args.path_connectomes
path_out = args.path_out
n_folds = args.n_folds
print('Loading data...')
pheno = pd.read_csv(path_pheno,index_col=0)
connectomes = pd.read_csv(path_connectomes,index_col=0)
print('Done!')
regressors_mc = ['AGE','C(SEX)','FD_scrubbed', 'C(SITE)', 'mean_conn']
#####################
# CONTINUOUS SCORES #
#####################
prs = ['Stand_PRS_height',
'Stand_PRS_BMI',
'Stand_PRS_BIP',
'Stand_PRS_newCDG2_ukbb',
'Stand_PRS_ASD',
'Stand_PRS_SCZ',
'Stand_PRS_LDL',
'Stand_PRS_CKD',
'Stand_PRS_IQ']
cont = prs + ['fluid_intelligence_score_all','Neuroticism']
cont_rows = []
for c in cont:
print(f'{c} - estimating effect size...')
p = pheno.copy()
if ('Stand_PRS' in c):
p = p[p['PI']=='UKBB']
p = p[(p.PRS_eth == 'WB') | (p.PRS_eth == 'EUR')]
if ('Stand_' not in c):
print('Dropping ',p[c].isna().sum(),' subjects w/ NaN for {}.'.format(c))
if (p[c].isna().sum() == p.shape[0]):
print('ERROR: No subjects with data for {}.'.format(c))
p = p.dropna(subset=[c])
print('Z-scoring contrast...')
p['{}_z'.format(c)] = (p[c] - p[c].mean())/p[c].std(ddof=0)
pheno['{}_z'.format(c)] = p['{}_z'.format(c)]
c = '{}_z'.format(c)
null_mask = np.array(~p[c].isnull())
p = p[null_mask]
mask = pheno.index.isin(p.index)
idx = pheno[mask].index
strat_col = np.ones(len(idx)) # DUMMY COL
skf = StratifiedKFold(n_splits=n_folds)
es_train = []
es_test = []
split_train_idx = []
split_test_idx = []
k=0
for train_index, test_index in skf.split(idx, strat_col):
print(f'Fold {k+1}/{n_folds}...')
# Save split indexes for the bootstrap
split_train_idx.append(idx[train_index].to_list())
split_train_idx.append(idx[test_index].to_list())
betamap_train = variable_effect(pheno.loc[idx[train_index]],
c,
regressors_mc,
connectomes.loc[idx[train_index]].to_numpy())['betas_std']
rank = pd.qcut(betamap_train.abs(),10,labels=False)
decile_idx = rank[rank==9].index
# Get train ES
decile_train = betamap_train.abs()[decile_idx]
mtd_train = np.mean(decile_train)
es_train.append(mtd_train)
betamap_test = variable_effect(pheno.loc[idx[test_index]],
c,
regressors_mc,
connectomes.loc[idx[test_index]].to_numpy())['betas_std']
# Get test ES
decile_test = betamap_test.abs()[decile_idx]
mtd_test = np.mean(decile_test)
es_test.append(mtd_test)
k += 1
print(f'{c} - CV Effect size ',np.mean(es_test))
cont_rows.append([c,np.mean(es_train),np.mean(es_test)])
df_split_idx = pd.DataFrame(zip(split_train_idx,split_train_idx),columns=['train','test'])
df_split_idx.to_csv(os.path.join(path_out,f'{c}_fold_idx.csv'))
print('Saving continuous results...')
df_cont = pd.DataFrame(cont_rows,columns=['case','train_ES','test_ES'])
df_cont.to_csv(os.path.join(path_out,'cont_CV_es.csv'))
print('Done!')