-
Notifications
You must be signed in to change notification settings - Fork 0
/
binomial.py
174 lines (106 loc) · 5.65 KB
/
binomial.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
import pandas as pd
import numpy as np
import math,os
from tqdm.auto import tqdm
from naming_scheme import type_and_suffix
def disease_pairs(obj:type_and_suffix,frequency=None,binomial=True):
from scipy.stats import binomtest
import warnings
warnings.simplefilter(action='ignore', category=pd.errors.PerformanceWarning)
os.makedirs('binomial',exist_ok=True)
(deaths,cox_index,frequency_default),for_cox=obj.read_cox_intermediate()
if frequency==None:
frequency=frequency_default
print(f'Frequency:{frequency}')
for_lr=pd.DataFrame()
base = pd.DataFrame(columns=['frequency','unorder_frequency','same_day_percentage'])
base.index.name='disease_pair'
p_unsatisfied=0
f_unsatisfied=0
for D1 in tqdm(cox_index):
if obj.disease_type=='death': D2s=deaths
else: D2s=cox_index
for D2 in D2s:
indexes=((for_cox[D1+'_ICD']== True) & (for_cox[D2+'_ICD']==True))
indexes=indexes&((for_cox[D1+'_in']== True) & (for_cox[D2+'_in']==True))
total=sum((for_cox.loc[indexes,D1+'_days']-for_cox.loc[indexes,D2+'_days'])!=0)
patients=(indexes & ((for_cox[D1+'_days']-for_cox[D2+'_days'])<0))
D1_first=sum(patients)
if D1_first>frequency:
if binomial==True:
threshold=0.05/D1_first
same_day=1-(total/sum(indexes))
bino=binomtest(D1_first,total,p=0.5, alternative='greater')
if bino.pvalue<threshold:
base.loc[D1+'_'+D2]=[D1_first,total,same_day]
for_lr[D1+'_'+D2]=patients.astype(int)
else:
p_unsatisfied+=1
else :
same_day=1-(total/sum(indexes))
base.loc[D1+'_'+D2]=[D1_first,total,same_day]
for_lr[D1+'_'+D2]=patients.astype(int)
else: f_unsatisfied+=1
print(f'satisfied\tpairs:{base.shape[0]}\n\
p_unsatisfied\tpairs:{p_unsatisfied}\n\
f_unsatisfied\tpairs:{f_unsatisfied}\n\
total \tpairs:{len(cox_index)*(len(D2s))}')
base.to_csv('binomial/binomial'+obj.get_suffix()+'.csv')
for_lr.to_csv('binomial/for_lr'+obj.get_suffix()+'.csv')
def disease_pairs_subgroup(obj:type_and_suffix):
from scipy.stats import binomtest
import warnings
warnings.simplefilter(action='ignore', category=pd.errors.PerformanceWarning)
_,for_cox=obj.read_cox_intermediate()
pairs=pd.read_csv('con_lr/T2'+obj.get_cox_suffix()+'.csv',index_col=0)
base = pd.DataFrame(columns=['frequency','unorder_frequency','same_day_percentage'])
base.index.name='disease_pair'
for Disease_pair in tqdm(pairs.index):
try:
[D1,D2]=Disease_pair.split(sep='_')
indexes=((for_cox[D1+'_ICD']== True) & (for_cox[D2+'_ICD']==True))
indexes=indexes&((for_cox[D1+'_in']== True) & (for_cox[D2+'_in']==True))
total=sum((for_cox.loc[indexes,D1+'_days']-for_cox.loc[indexes,D2+'_days'])!=0)
patients=(indexes & ((for_cox[D1+'_days']-for_cox[D2+'_days'])<0))
D1_first=sum(patients)
same_day=1-(total/sum(indexes))
base.loc[D1+'_'+D2]=[D1_first,total,same_day]
except:
base.loc[D1+'_'+D2]=[0,0,0]
base.to_csv('binomial/binomial'+obj.get_suffix()+'.csv')
def disease_death_pairs(obj:type_and_suffix,frequency=None):
from scipy.stats import binomtest
import warnings
warnings.simplefilter(action='ignore', category=pd.errors.PerformanceWarning)
(_,cox_index,frequency_default),for_cox=obj.read_cox_intermediate()
if frequency==None:
frequency=frequency_default
for_lr = pd.DataFrame()
base = pd.DataFrame(columns=['frequency','unorder_frequency','same_day_percentage'])
base.index.name='disease_pair'
p_unsatisfied=0
f_unsatisfied=0
for D1 in tqdm(cox_index):
for D2 in cox_index:
indexes=((for_cox[D1+'_ICD']== True) & (for_cox[D2+'_ICD']==True)& (for_cox[obj.Death_cause+'_ICD']==True))
indexes=indexes&((for_cox[D1+'_in']== True) & (for_cox[D2+'_in']==True))
total=sum((for_cox.loc[indexes,D1+'_days']-for_cox.loc[indexes,D2+'_days'])!=0)
patients=(indexes &\
((for_cox[D1+'_days']-for_cox[D2+'_days'])<0)& \
((for_cox[D2+'_days']-for_cox[obj.Death_cause+'_days'])<0))
D1_first=sum(patients)
if D1_first>frequency:
threshold=0.05/D1_first
same_day=1-(total/sum(indexes))
bino=binomtest(D1_first,total,p=0.5, alternative='greater')
if bino.pvalue<threshold:
base.loc[D1+'_'+D2]=[D1_first,total,same_day]
for_lr[D1+'_'+D2]=patients.astype(int)
else:
p_unsatisfied+=1
else: f_unsatisfied+=1
print(f'satisfied\tpairs:{base.shape[0]}\n\
p_unsatisfied\tpairs:{p_unsatisfied}\n\
f_unsatisfied\tpairs:{f_unsatisfied}\n\
total \tpairs:{len(cox_index)*(len(cox_index)-1)}')
base.to_csv('binomial/binomial'+obj.get_suffix()+'.csv')