utility/flatten_multi_records.py at main · AMP-SCZ/utility · GitHub

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
#!/usr/bin/env python

import pandas as pd
import sys
import re
from os.path import isfile

# useful settings for debugging
pd.set_option("display.max_rows", None)
pd.set_option("display.max_columns", None)
pd.set_option("display.max_colwidth", None)


multi_records= {'adverse_events': {'visit': 99, 'vars': ['chrae_aescreen', 'chrae_?', 'chrae_tp?', 'chrae_diag?', 'chrae_ae?', 'chrae_aes?date', 'chrae_aes?off', 'chr_ae?date', 'chrae_sig?', 'chrae_dr?', 'chrae_d?', 'chrae_e?', 'chrae_expected?', 'chrae_sae?', 'chrae_ssi?', 'chr_ae?date_dr', 'chrae_sig?_dr', 'chrae_ae?_trans_q', 'chrae_ae?_mo1', 'chrae_ae?_mo2', 'chrae_ae?_mo3', 'chrae_ae?_mo4', 'chrae_ae?_mo5', 'chrae_ae?_mo6', 'chrae_ae?_mo7', 'chrae_ae?_mo8', 'chrae_ae?_mo9', 'chrae_ae?_mo10', 'chrae_ae?_mo11', 'chrae_ae?_mo12', 'chrae_ae?_mo18', 'chrae_ae?_mo24', 'chrae_ae?_trans', 'chrae_ae?_offmes', 'chrae_ae?_comments','chrae_add?']},

'psychosocial_treatment_form': {'visit': 99, 'vars': ['chrpsychsoc_concerns', 'chrpsychsoc_treat?', 'chrpsychsoc_treat?_tp', 'chrpsychsoc_treat?_type', 'chrpsychsoc_treat?_other', 'chrpsychsoc_treat?_res', 'chrpsychsoc_treat?_freq', 'chrpsychsoc_treat?_onset', 'chrpsychsoc_treat?_offset', 'chrpsychsoc_treat?_instruct', 'chrpsychsoc_treat?_instruct_other', 'chrpsychsoc_treat?_trans_q', 'chrpsychsoc_treat?_mo1', 'chrpsychsoc_treat?_mo2', 'chrpsychsoc_treat?_mo3', 'chrpsychsoc_treat?_mo4', 'chrpsychsoc_treat?_mo5', 'chrpsychsoc_treat?_mo6', 'chrpsychsoc_treat?_mo7', 'chrpsychsoc_treat?_mo8', 'chrpsychsoc_treat?_mo9', 'chrpsychsoc_treat?_mo10', 'chrpsychsoc_treat?_mo11', 'chrpsychsoc_treat?_mo12', 'chrpsychsoc_treat?_mo18', 'chrpsychsoc_treat?_mo24', 'chrpsychsoc_treat?_trans', 'chrpsychsoc_treat?_offmes', 'chrpsychsoc_treat?_comments','chrpsychsoc_treat?_add']},

'health_conditions_medical_historypsychiatric_histo': {'visit': 99, 'vars': ['chrmed_any', 'chrmed_cond?', 'chrmed_cond?_tp', 'chrmed_cond?_name', 'chrmed_cond?_onset', 'chrmed_cond?_offset', 'chrmed_cond?_instruct', 'chrmed_cond?_trans_q', 'chrmed_cond?_mo1', 'chrmed_cond?_mo2', 'chrmed_cond?_mo3', 'chrmed_cond?_mo4', 'chrmed_cond?_mo5', 'chrmed_cond?_mo6', 'chrmed_cond?_mo7', 'chrmed_cond?_mo8', 'chrmed_cond?_mo9', 'chrmed_cond?_mo10', 'chrmed_cond?_mo11', 'chrmed_cond?_mo12', 'chrmed_cond?_mo18', 'chrmed_cond?_mo24', 'chrmed_cond?_trans', 'chrmed_cond?_offmes', 'chrmed_cond?_comments','chrmed_cond?_add']},

'resource_use_log': {'visit': 99, 'vars': ['chrrul_any', 'chrrul_resource?', 'chrrul_resource?_tp', 'chrrul_resource?_code', 'chrrul_resource?_on', 'chrrul_resource?_off', 'chrrul_resource?_notes','chrrul_resource?_add']},

'traumatic_brain_injury_screen_subject': {'visit': 1, 'vars': ['chrtbi_subject_age?', 'chrtbi_subject_circumstance?', 'chrtbi_subject_length?', 'chrtbi_subject_anterograde?', 'chrtbi_subject_retrograde?', 'chrtbi_subject_symptoms2_?', 'chrtbi_subject_symptom_length?', 'chrtbi_subject_med_find?']},
'traumatic_brain_injury_screen_parent': {'visit': 1, 'vars': ['chrtbi_parent_age?', 'chrtbi_parent_circumstance?', 'chrtbi_parent_length?', 'chrtbi_parent_anterograde?', 'chrtbi_parent_retrograde?', 'chrtbi_parent_symptoms2_?', 'chrtbi_parent_symptom_length?', 'chrtbi_parent_medical_findings?']},

'blood_sample_preanalytic_quality_assurance_bc': {'visit': [2,4], 'vars': ['chrblood_bc?id','chrblood_bc?vol','chrblood_bc?pos','chrblood_bc?box','chrblood_buffy_freeze','chrblood_bcfrztime']},
'blood_sample_preanalytic_quality_assurance_pl': {'visit': [2,4], 'vars': ['chrblood_pl?id','chrblood_pl?vol','chrblood_pl?pos','chrblood_pl?hem','chrblood_plasma_freeze','chrblood_plfrztime']},
'blood_sample_preanalytic_quality_assurance_wb': {'visit': [2,4], 'vars': ['chrblood_wb?id','chrblood_wb?vol','chrblood_wb?pos','chrblood_wholeblood_freeze','chrblood_wbfrztime']},
'blood_sample_preanalytic_quality_assurance_se': {'visit': [2,4], 'vars': ['chrblood_se?id','chrblood_se?vol','chrblood_se?pos','chrblood_serum_freeze','chrblood_serumfrztime','chrblood_sehem','chrblood_selip']},

'family_interview_for_genetic_studies_figs_child': {'visit': 1, 'vars': ['chrfigs_child?_age','chrfigs_child?_sex','chrfigs_child?_info','chrfigs_child?_s01','chrfigs_child?_s02','chrfigs_child?_s03','chrfigs_child?_s04','chrfigs_child?_s05','chrfigs_child?_s06','chrfigs_child?_s07','chrfigs_child?_s08','chrfigs_child?_d01','chrfigs_child?_d02','chrfigs_child?_d03','chrfigs_child?_d04','chrfigs_child?_d05','chrfigs_child?_d06','chrfigs_child?_d07','chrfigs_child?_d08','chrfigs_child?_d10','chrfigs_child?_d11','chrfigs_child?_d12','chrfigs_child?_m01','chrfigs_child?_m02','chrfigs_child?_m03','chrfigs_child?_m04','chrfigs_child?_m05','chrfigs_child?_m06','chrfigs_child?_m07','chrfigs_child?_m08','chrfigs_child?_m09','chrfigs_child?_m10','chrfigs_child?_m11','chrfigs_child?_m12','chrfigs_child?_p01','chrfigs_child?_p02','chrfigs_child?_p03','chrfigs_child?_p04','chrfigs_child?_p05','chrfigs_child?_p06','chrfigs_child?_p07','chrfigs_child?_p08','chrfigs_child?_p09','chrfigs_child?_p10','chrfigs_child?_p11','chrfigs_child?_p12','chrfigs_child?_p13','chrfigs_child?_p14','chrfigs_child?_p15','chrfigs_child?_p16','chrfigs_child?_p17','chrfigs_child?_p18','chrfigs_child?_ddx','chrfigs_child?_ddx_agree','chrfigs_child?_ddx_why','chrfigs_child?_mdx','chrfigs_child?_mdx_agree','chrfigs_child?_mdx_why','chrfigs_child?_pdx','chrfigs_child?_pdx_agree','chrfigs_child?_pdx_why','chrfigs_child?_napdx','chrfigs_child?_napdx_agree','chrfigs_child?_napdx_why','chrfigs_child?_apdx','chrfigs_child?_apdx_agree','chrfigs_child?_apdx_why']},
'family_interview_for_genetic_studies_figs_sibling': {'visit': 1, 'vars': ['chrfigs_sibling?_age','chrfigs_sibling?_sex','chrfigs_sibling?_info','chrfigs_sibling?_s01','chrfigs_sibling?_s02','chrfigs_sibling?_s03','chrfigs_sibling?_s04','chrfigs_sibling?_s05','chrfigs_sibling?_s06','chrfigs_sibling?_s07','chrfigs_sibling?_s08','chrfigs_sibling?_d01','chrfigs_sibling?_d02','chrfigs_sibling?_d03','chrfigs_sibling?_d04','chrfigs_sibling?_d05','chrfigs_sibling?_d06','chrfigs_sibling?_d07','chrfigs_sibling?_d08','chrfigs_sibling?_d10','chrfigs_sibling?_d11','chrfigs_sibling?_d12','chrfigs_sibling?_m01','chrfigs_sibling?_m02','chrfigs_sibling?_m03','chrfigs_sibling?_m04','chrfigs_sibling?_m05','chrfigs_sibling?_m06','chrfigs_sibling?_m07','chrfigs_sibling?_m08','chrfigs_sibling?_m09','chrfigs_sibling?_m10','chrfigs_sibling?_m11','chrfigs_sibling?_m12','chrfigs_sibling?_p01','chrfigs_sibling?_p02','chrfigs_sibling?_p03','chrfigs_sibling?_p04','chrfigs_sibling?_p05','chrfigs_sibling?_p06','chrfigs_sibling?_p07','chrfigs_sibling?_p08','chrfigs_sibling?_p09','chrfigs_sibling?_p10','chrfigs_sibling?_p11','chrfigs_sibling?_p12','chrfigs_sibling?_p13','chrfigs_sibling?_p14','chrfigs_sibling?_p15','chrfigs_sibling?_p16','chrfigs_sibling?_p17','chrfigs_sibling?_p18','chrfigs_sibling?_ddx','chrfigs_sibling?_ddx_agree','chrfigs_sibling?_ddx_why','chrfigs_sibling?_mdx','chrfigs_sibling?_mdx_agree','chrfigs_sibling?_mdx_why','chrfigs_sibling?_pdx','chrfigs_sibling?_pdx_agree','chrfigs_sibling?_pdx_why','chrfigs_sibling?_napdx','chrfigs_sibling?_napdx_agree','chrfigs_sibling?_napdx_why','chrfigs_sibling?_apdx','chrfigs_sibling?_apdx_agree','chrfigs_sibling?_apdx_why']},

'current_pharmaceutical_treatment_floating_med_125': {'visit': 99, 'vars': ['chrpharm_med','chrpharm_med?_comments','chrpharm_med?_name','chrpharm_dose_lowerlimit_med?','chrpharm_dose_upperlimit_med?','chrpharm_med?_onset','chrpharm_med?_offset','chrpharm_firstdose_med?','chrpharm_med?_use','chrpharm_med?_dosage','chrpharm_med?_frequency','chrpharm_med?_datasource','chrpharm_med?_other','chrpharm_med?_comp','chrpharm_med?_indication','chrpharm_med?_other2','chrpharm_med?_mo0','chrpharm_med?_mo1','chrpharm_med?_mo2','chrpharm_med?_mo3','chrpharm_med?_mo4','chrpharm_med?_mo5','chrpharm_med?_mo6','chrpharm_med?_mo7','chrpharm_med?_mo8','chrpharm_med?_mo9','chrpharm_med?_mo10','chrpharm_med?_mo11','chrpharm_med?_mo12','chrpharm_med?_mo18','chrpharm_med?_mo24','chrpharm_med?_add']},
'current_pharmaceutical_treatment_floating_med_2650': {'visit': 99, 'vars': ['chrpharm_med?_comments','chrpharm_med?_name','chrpharm_dose_lowerlimit_med?','chrpharm_dose_upperlimit_med?','chrpharm_med?_onset','chrpharm_med?_offset','chrpharm_firstdose_med?','chrpharm_med?_use','chrpharm_med?_dosage','chrpharm_med?_frequency','chrpharm_med?_datasource','chrpharm_med?_other','chrpharm_med?_comp','chrpharm_med?_indication','chrpharm_med?_other2','chrpharm_med?_mo0','chrpharm_med?_mo1','chrpharm_med?_mo2','chrpharm_med?_mo3','chrpharm_med?_mo4','chrpharm_med?_mo5','chrpharm_med?_mo6','chrpharm_med?_mo7','chrpharm_med?_mo8','chrpharm_med?_mo9','chrpharm_med?_mo10','chrpharm_med?_mo11','chrpharm_med?_mo12','chrpharm_med?_mo18','chrpharm_med?_mo24','chrpharm_med?_add']},

'past_pharmaceutical_treatment': {'visit': 1, 'vars': ['chrpharm_med_past','chrpharm_med?_comments_past','chrpharm_med?_name_past','chrpharm_dose_lowerlimit_med?_past','chrpharm_dose_upperlimit_med?_past','chrpharm_med?_onset_past','chrpharm_med?_offset_past','chrpharm_med?_use_past','chrpharm_med?_dosage_past','chrpharm_med?_frequency_past','chrpharm_med?_datasource_past','chrpharm_med?_other_past','chrpharm_med?_comp_past','chrpharm_med1_indication_past','chrpharm_med?_other2_past','chrpharm_med?_add_past']},

}


def flatten_group(df,form,cols):

    dict1={}
    dict1['visit']= df.loc[0,'visit'] if 'visit' in df else multi_records[form]['visit']
    for v in multi_records[form]['vars']:

        # has the variable been exported by RPMS?
        if v.replace('?','') in cols:
            c= v.replace('?','')
        elif v.replace('_?','') in cols:
            c= v.replace('_?','')
        else:
            continue

        # go through the rows of that variable and generate a flat list
        for i,row in df.iterrows():
            if '?' in v:
                dict1[v.replace('?',str(row['Row#']))]= row[c]

            elif i==0:
                # some variables like
                # chrblood_bcfrztime, chrblood_plfrztime, chrblood_wbfrztime, chrblood_serumfrztime
                # have valid values only in the first row
                dict1[v]= row[c]


    # for n repeats, there are n-1 *_add vars, so delete the nth *_add var
    for v in multi_records[form]['vars']:
        if '_add' in v:
            last= df.loc[df.shape[0]-1,'Row#']
            del dict1[v.replace('?',last)]
            break

    return dict1


def flatten_one_new(filename):

    dfmulti= pd.read_csv(filename,dtype=str,keep_default_na=False)

    subjectkey= filename.split('_')[0]
    form= re.search(f'{subjectkey}_(.+?).csv', filename).group(1)

    # default_cols= ['LastModifiedDate','subjectkey','interview_date','interview_age','gender']
    # interview_date variable has become different across forms
    # so use the following for loop to obtain default columns
    default_cols=[]
    for c in dfmulti.columns:
        default_cols.append(c)
        if c=='gender':
            break

    cols= [c for c in dfmulti.columns if c not in default_cols]

    dfall= pd.DataFrame()
    groups= dfmulti.groupby('visit')
    for timepoint in groups.groups.keys():

        dfvisit= groups.get_group(timepoint).reset_index()

        # new single-row data frame with default columns for that timepoint/visit
        df1= dfvisit.loc[:0][default_cols]
        dict1= flatten_group(dfvisit,form,cols)

        # vertically concatenate default columns and flat list
        df1= pd.concat([df1,pd.DataFrame([dict1])], axis=1)

        # horizontally concatenate flat lists
        dfall= pd.concat([dfall,df1], axis=0, sort=False)


    return dfall


def flatten_many_new():

    filename=sys.argv[1]

    dfunique= pd.read_csv(filename,dtype=str,keep_default_na=False)
    subjectkey= filename.split('_')[0]

    # default_cols= ['LastModifiedDate','subjectkey','interview_date','interview_age','gender','visit']
    # interview_date variable has become different across forms
    # so use the following for loop to obtain default columns
    default_cols=[]
    for c in dfunique.columns:
        default_cols.append(c)
        if c=='visit':
            break

    dfall= pd.DataFrame()
    for i,row in dfunique.iterrows():

        timepoint=row['visit']

        df1= pd.DataFrame(columns=dfunique.columns)
        df1.loc[0]= row

        dict1={}
        for filename in sys.argv[2:]:

            if not isfile(filename):
                continue

            dfmulti=pd.read_csv(filename,dtype=str,keep_default_na=False)
            groups= dfmulti.groupby('visit')
            dfvisit= groups.get_group(timepoint).reset_index()

            form= re.search(f'{subjectkey}_(.+?).csv', filename).group(1)
            cols= [c for c in dfmulti.columns if c not in default_cols]
            _dict1= flatten_group(dfvisit,form,cols)
            # visit column is inherited from unique record, so omit its possible repetition
            del _dict1['visit']

            # vertically concatenate flat lists across split files
            dict1.update(_dict1)


        # vertically concatenate default columns and flat list
        df1= pd.concat([df1,pd.DataFrame([dict1])], axis=1)

        # horizontally concatenate flat lists
        dfall= pd.concat([dfall,df1], axis=0, sort=False)


    return dfall


if __name__=='__main__':

    if len(sys.argv)<2 or sys.argv[1] in ['-h','--help']:
        print(f'''Usage: {__file__} /path/to/ME57953_multi_record_form.csv
This program transforms a multi-row record into uniquely-named columns for REDCap import.
Observe *.csv.flat output file.''')
        exit(0)

    if not isfile(sys.argv[1]):
        print(f'unique-record file {sys.argv[1]} does not exist')
        exit()

    if len(sys.argv)==2:
        df2=flatten_one_new(sys.argv[1])
    else:
        df2=flatten_many_new()

    output= sys.argv[1]+'.flat'
    if len(df2):
        df2.to_csv(output, index=False)
        print('Generated', output)