Ontology meeting 2018-07-02

From GO Wiki
Jump to navigation Jump to search

Conference Line

Agenda

Editors Discussion

Project Update

ECM Project--- https://github.com/geneontology/go-ontology/projects/9

Regenerate Imports File

Erich, David, Kimberly and Pascale loaded Docker and tried to regenerate the cell ontology import but were unsuccessful.

Erich think he has solved the issue. We will try again this week.

Obsoletion plugin for Protege

Please add you Protege versions and the plugin you have here.

Barbara- Protege 5.2.0
David- Protege 5.1.0, OBO annotation plugin 0.4.0
Harold- Protege 5.1.0, OBO annotation plugin 0.4.0
Karen- Protege 5.1.0, OBO Annotations Editor 0.4.0
Kimberly- Protege 5.1.0, I have something called OBO annotation editors 0.3.0 - is that the plugin?

Ontology Documentation

Pascale, David and Kimberly have been working on standardizing the ontology documentation to make it align with the documentation work that Kimberly and Pascal have been doing for other areas of GO. Specifically, we have moved the official documentation to the wiki and have arranged the wiki in a way that is similar to the annotation section. We are now going through the documentation to check it for accuracy and to be sure it is up to date.

Last week Pascale and David worked on the definitions section.


Ontology Developers' workshop possibility

  • Push to November

GH project link

https://github.com/geneontology/go-ontology/projects/1

Minutes

  • On call: Chris, David, Eric, Harold, Jim, Judy, Karen, Kimberly, Laurent-Philippe, Pascale, Peter, Rachael, Tanya, Terry

Project Update

ECM Project--- https://github.com/geneontology/go-ontology/projects/9

  • Some of the existing terms, e.g. lamina reticularis, are now believed to correspond to experimental artifacts
  • Some external ontologies, e.g. mouse anatomy ontology and UBERON, still use these terms, however
  • What should GO do in these cases?
  • These terms are on the bridge between subcellular terms and anatomical terms
  • Do the anatomists use these terms the same way that GO would use them?
    • GO has these terms as parts of (e.g. layers) of basement membranes
  • In these cases, we should put an obsoletion notice on terms to solicit feedback
    • This would be a new annotation property that could be used to generate reports about pending obsoletions
    • Any term slated for obsoletion would ge this tag
    • GO editors would also need to use their judgement about announcing these changes more widely
    • Ideally, this system would be employed across all OBO ontologies
  • David did send an obsoletion notice, but when looking in Ontobee didn't see that the term was used by any of the anatomical dictionaries
  • Should we resurrect the term?
    • For now, we will resurrect the lamina reticularis term but note that it hasn't been used for annotation and, in one camp, seems to be considered artifacts
  • AI: Rachael will start an email thread including the experts from the various disciplines to get everyone with a vested interest on the same communication
  • AI: Pascale will create the github/notifications repo
  • AI: David and Rachael will resurrect the lamina reticularis term on Friday

Regenerate Imports File

  • David, Eric, Kimberly, Pascale will try again
  • Installed Docker, tried but ran into a snag; will try again today (July 2nd)

Obsoletion plugin for Protege

  • Will all try to get the same plugin; use that for obsoletions

Ontology Editors Meeting

  • Denver the last week of November, before the Alliance all-hands meeting

Aligning GO-Rhea-Reactome

  • Updating the GO-Reactome xrefs to only xref to the human Reactome reactions
    • Found some false negatives, but these are likely due to GO not having updated Reactome xrefs in quite some time
  • Updating the GO-Rhea xrefs to the Rhea master xref, as opposed to the bidirectional child reaction
    • This update is fine with Rhea, too
  • Aiming for a merge of these xrefs this afternoon
  • Updating logical definitions for enzymatic activity terms using all chemical participants
    • There will be a new ChEBI import file that has all of the chemicals required for representing the participants
    • There may be an issue with file size in github, but the reasoner should be fine
    • Will need to see, in practice, how large the actual new import file is
    • In the obo file, we will lose the nested nature of the participants list, but does this matter if we never edit them?
    • Should we abandon obo in place of owl?
      • We don't need to do this right now
      • The Rhea imports will exist as a separate file, editors will be able to see them in Protege, but would not edit them
      • Any issues would need to be addressed with Rhea, ChEBI, etc.; basically the pipelines or ontologies upstream of GO
  • David will give a more detailed presentation on this next week

Individual github Tickets

  • There are still a number of tickets that have the 'Editors Discussion' label on them
  • If this needs to be discussed on a call, please let David know so it can be prioritized for the Monday call
  • For 'only in' taxon constraints, editors need only add the non-redundant taxa