QA of DrugCentral Ingest

[mbrush]

QA Resources:

• KGX Summary report: https://docs.google.com/spreadsheets/d/1vHAvBJPU4c72cZR1d7Dr2-Z1Dt9epTZ-pXSkfOp3wqw/edit?gid=894378999#gid=894378999
• RIG: https://github.com/NCATSTranslator/translator-ingests/blob/main/src/translator_ingest/ingests/drugcentral/drugcentral_rig.yaml
• RIG updates PR: n/a

Issues to consider / address now:
Overall ingest and RIG look good. Just one minor point to consider below. And @colleenXu if you haven't yet reviewed the KGX Summary Report here to make sure that the edge types and counts seem reasonable to you, please do. Thanks!

• Noting that we use/populate the clinical_approval_status edge property on treats edges only in the very small subset that describe off-label use. Do we know that all other treats edges are for on-label / approved drugs, so we can report their clinical_approval_Status as "approved"? FOr the same of completeness / symmetry here.

---

Issues to consider longer term:

• see future considerations in RIG.

---

Questions:

[colleenXu]

@mbrush

I do not know for sure whether the other terms mapped to treats (indication, symptomatic treatment) are for on-label/approved use. But I think it's reasonable to add clinical_approval_status: approved_for_condition to the indication edges, which is almost all of the other treats edges.

Reasoning:
The resource table does not include approval status - only the term name.
Reading between the lines here, the DrugCentral does seem to equate indication with on-label/approved use. But they don't explicitly mention other terms like symptomatic treatment or reduce risk (mapped to biolink preventative_for_condition), so I have no idea what the approval status would be for those edges.
But there's only 1-2 rows in the resource table for symptomatic treatment and reduce risk, so I imagine it's fine we don't include the approval status info on those edges.

[mbrush]

I agree @colleenXu . . . make it so!