final steps regarding subject info

[yibeichan]

I finished the mapping task here

• for any given nimh-minimal reproschema response folder (downloaded from reproschema UI), we can convert them into CDE submission template.
• I separate the task into two major parts: (1) match ids based on questions and (2) map responses. each part I added a tiny agent to help. but agent is my last resort, only when regular matching/mapping methods not working, then our agents will work
• I also created enough tests and made ollama running on GA

questions about subject info (need @satra)

• in all submission templates, it requires subjectkey, src_subject_id, interview_age, interview_date, sex, but we only gave interview_age and sex collected. here you can find their definition, our NDA reproschema library collection is not updated. I guess we want to update them?
• so we will directly add src_subject_id and interview_date to nimh-minimal then?
• for subjectkey, which is "The NDAR Global Unique Identifier (GUID) for research subject", daniel gave me a script last year to "Generate a unique URL (for users or testing)" I guess we will use it to get GUID?

anything else you expect to be done for this mapping task? i guess we also need to move this repo to repronim, also, do we want it to be an independent repo or do we want to integrate it into reproschema-py?

[satra]

interview_date should come from the response timestamp and subjectkey should be an ndar guid created using their service. this would be attached to url using uid=<guid>. src_subject_key should be the entity id for the agent generated automatically for the response.

[yibeichan]

interview_date yes we can automatically generate from response, but then why do we need this one https://github.com/yibeichan/reproschema-library/blob/main/activities/NDA/items/interview_date?

src_subject_key this one yes we can prompt agent to do it, maybe we should add an argument, since different studies might have different naming criteria and we may want all participants from the same study follow the same naming convention?

subjectkey now I'm confused

should be an ndar guid created using their service. this would be attached to url using uid=

do we do anything for this one or not?

[satra]

it may be helpful to add a subject id field to the demographics form. this can be the same as uid if it was provided, or the uid can be set to this value. both can be used to populate the subjectkey and/or src_subject_key

[yibeichan]

okay, understood.

one more: do we want to update our NDA reproschema library collection with NDA CDE here. if not urgent, we don't have to do it (I can go back to reproschema-py to fix our converter)

[satra]

nice to do, but not urgent.

[yibeichan]

okay, I added interview_date (automatically extracted) and src_subject_id from the schema; worked well.