The function “check_subj” has a parameter of “subj_exp” which is to be a dataframe with two variables (“SUBJECT_ID”,”CONSENT”). The code then compares the consent in this set with consent variable in the subject-consent dbgap file. Comparing consents I think needs manual intervention so I’m not sure what the best thing would be for any modification of “check_subj”.
Example: My understanding is that “subj_exp” would usually be the subjects in the freeze that we are currently working on. However the consent code information in the master subject annotation is in a different format than the CONSENT variable in the subject file (e.g. for SAFS, consent in subject file is 0 or 1, with the definition of 1 being in the data dictionary; the consent in the subject annotation is “DS-DHD-IRB-PUB-MDS-RD”. It does ‘match’ the definition in the subject file data dictionary but ….
This could be addressed by substituting the consent code for the value in the data dictionary, and then doing a table of expected value vs observed value instead of reporting mismatches.
The function “check_subj” has a parameter of “subj_exp” which is to be a dataframe with two variables (“SUBJECT_ID”,”CONSENT”). The code then compares the consent in this set with consent variable in the subject-consent dbgap file. Comparing consents I think needs manual intervention so I’m not sure what the best thing would be for any modification of “check_subj”.
Example: My understanding is that “subj_exp” would usually be the subjects in the freeze that we are currently working on. However the consent code information in the master subject annotation is in a different format than the CONSENT variable in the subject file (e.g. for SAFS, consent in subject file is 0 or 1, with the definition of 1 being in the data dictionary; the consent in the subject annotation is “DS-DHD-IRB-PUB-MDS-RD”. It does ‘match’ the definition in the subject file data dictionary but ….