Create a report to evaluate obsolete replacement relations

[sabrinatoro]

When we update obsolete references in external resources, we should keep track of the following information:

• reasons for deprecation (ie why did the external resource obsolete the term?)
• term to consider
• “replaced by” term
  This information will help us determine whether terms in Mondo should also be obsoleted or merged.
  Until we had a few round of manual reviews (and find patterns for each resources), no changes should be done directly to Mondo.

I don’t know what information will be available from the external sources, and I am certain that all external sources have different information available.
Please create a report with as much of the following information as possible.

• External source ID
• Mondo ID (based on Mondo:equivalentTo)
• Reason for obsoletion in the external source
• Term to consider - external source ID (from the external source)
• Term to consider - Mondo ID
  • = Mondo term corresponding to the term to consider external source ID; based on mapping/Mondo:equivalentTo
• “Replaced by” - external source ID (from the external source)
  • this is the term replacing the now-obsoleted term according to the external source
  • in OMIM, they use the words “moved to” (e.g ^100500 MOVED TO 200150
• “Replaced by” - Mondo ID
  • = Mondo term corresponding to the “replaced by” in the external source ID; based on mapping/Mondo:equivalentTo)

Please create this report to include all the xref that have the source "Mondo:equivalentobsolete".
Moving forward, please create this report every time the update obsolete references in external resources pipeline run

Note: let me know if you want me to review the files from the external source and advice on what should be “term to consider” or “term replaced by”.

[sabrinatoro]

@matentzn I created this issue based on our last conversation.

[matentzn]

I think this should be moved to mondo ingest repo and described as another sync task. In the context of the Mondo repo we dont have access to the raw source data.

The OP needs to be amended a bit:

• ensure that all sources are screened for variants of the "replaced-by" relation and mapped to IAO:replacedBy (find correct ID).

Then it is really about creating the report you suggest using a python script from the maybe a SPARQL generated table of all replaced by relations in the sources.