Several updates (morbid map, labeling, genes)

[joeflack4]

Updates

• Bugfix: morbidmap.txt parsing: Fixing various issues related to how we're parsing this and adding to the graph, now that we've discovered that several assumptions are false (e.g. oftentimes there is no phenotype MIM number).
• Updates: morbidmap.txt parser:
  • Now extracting phenotype mapping / association evidence information.
  • Also reified the way we're representing these relationships. No longer a single triple using a single RO relationship.
  • We're using other RO predicates now as well.
  • In the case of multiple mapping evidence, we're just using RO:0003302 (https://github.com/monarch-initiative/omim/issues/75).
  • In cases where no MIM# in Phenotype field, not adding an association.
  • In cases where the mapping key within the Phenotype field is '1', not adding an association (see: https://github.com/monarch-initiative/omim/issues/79#issuecomment-1319408780)
• Bugfix: In some cases when there was not a label / alt label, 'None' was being added as the label.I
• Bugfix: Invalid xrefs were being added, e.g. OMIM:1 -> OMIM:14. This was an artefact of inconsistency in intra-label data model within morbidmap.txt's "Phenotype" field; that is, sometimes there was a missing MIM number, and in those cases, if there was an integer at the end of the phenotype label, it was being interpreted as a MIM number.
• Misc: (i) added some todo's, (ii) code comments, (iii) minor refactorings.
• Update: Added CONFIG var with 'verbose' as an option, defaulting to False. Now prints no INFO unless this is True.
• Add: additional report TSV specific to morbidmap.txt Phenotypes with no MIM#
• Add: rdfs:comment regarding the text ', INCLUDED' which appears in some labels.
• Rename: data/omim.ttl --> data/legacy_omim.ttl

Addresses

• [x] #74
• [x] #67
• [x] #58
• [x] #8
• [x] #75
• [x] #78
• [ ] #79
  • This one may remain open afterwards due to some follow-up considerations.

Remaining sub-issues (#73)

• [x] 1. How to add more info about mapping evidence
• [x] 2. Sometimes I shouldn't add association evidence triple?

[matentzn]

@joeflack4 anything you need from me here?

[joeflack4]

@matentzn I think not. After our discussion on Friday and in this PR, I've added the following issues to be addressed by this PR (I've added a full list of related issues in the OP and in "Development" on the right-hand side of the PR page):

• 8

• 78

I've put this back to a draft for now. My goal is to complete these by Friday and have you review again.

[joeflack4]

@matentzn Ready again for review. omim.ttl.zip

[joeflack4]

@matentzn This has been ready to merge for a while. I just went through again and verified evrything resolved.

[joeflack4]

@matentzn i just looked at my notes and i saw you'd already given me approval to merge this after doing an analysis on obsolete terms, which was successfully completed