Review of Evidence Capture for Monarch Ingests

[RichardBruskiewich]

The Monarch Data meeting of 22nd June 2023, reviewed the need for quality assurance in the capture of Monarch graph edge evidence. We wish to:

• Clarify how to apply the current Biolink Model (Translator) standards for the capture and annotation of evidence, e.g. annotation of knowledge types and agents; capture of supporting citations
• Review every Monarch Ingest source for available metadata which can be used to populate the Biolink Model evidence model evidence slot values needed:
  • alliance
  • Modularize and update BGee ingest #843
  • ctd
  • dictybase
  • flybase
  • goa
  • hgnc
  • hpoa
  • mgi
  • ncbi
  • panther
  • pombase
  • reactome
  • rgd
  • sgd
  • string
  • xenbase
  • zfin

[RichardBruskiewich]

@kevinschaper is this issue already covered somewhere else (maybe @amc-corey-cox is doing most of the needful here?)

[amc-corey-cox]

@RichardBruskiewich I created a Jupyter notebook to look at the evidence in our final QC graph. I haven't dug into anything in the ingests. We were mostly concerned with evidence types and coverage for QC. I think this is tangential to what I'm working on now. With that said... I will try to get some insight on this as well.

[RichardBruskiewich]

Let me know if you need any Biolink Model backup on this issue (could potentially frame as a Translator outreach?)

[monicacecilia]

Hi @amc-corey-cox, did you make any other progress on this back then?
We've made progress in displaying evidence in the results tables. Are there any other improvements? If so, what?
@kevinschaper 👀 👆