search

ohdsi-studies / TkrPredictSimple

0 stars 1 forks source link

Trying to run this study package on Stanford~STARR OMOP #2

Open ablack3 opened 1 year ago

ablack3 commented 1 year ago

Hi all,

I'm working with a great team to run several OHDSI network studies on Stanford's CDM. I'll put my updates in this thread for record keeping. One initial question I have is what version of PLP should be used?

jreps commented 1 year ago

Looks like it may need an older version and that may be difficult to load given the R dependencies may have changed in the past 2 years. This is the problem of us not using renv in older studies. If it doesn't run with the current PLP I should be able to create a Strategus script to run this (and any other OHDSI prediction you want to run) fairly quickly (I could make it next Friday).

ablack3 commented 1 year ago

Thanks @jreps! Let me try to get this to run as is and maybe create a lockfile for it. First issue when trying to install this package is that Depends is used twice in the Description file causing an error.

image

https://github.com/ohdsi-studies/TkrPredictSimple/blob/eac112daccb1c34478fe18ea693f34ee8f6a7dee/DESCRIPTION#L10

ablack3 commented 1 year ago

Can I have access to this repo so I can make changes on a new branch?

jreps commented 1 year ago

I guess that trumps a lack of renv :) I've added you - you should get an email inviting you to this repo.

ablack3 commented 1 year ago

Ok thanks. Yea we are kind of trying to demonstrate replicability (same code, new data). If you write a strategus version I can try that as well or a more recent strategus study.

The cohorts generated using the most recent Hades packages so that's a good start.