Ontology meeting 2018-10-01

From GO Wiki
Jump to navigation Jump to search

Conference Line

Agenda

Meetings

Montreal GOC Meeting

Ontology Developers' workshop

Fall 2018 ontology editors meeting - potential agenda topics

Editors Discussion

Regulation inference chain

For Molecular Function, the 'regulates' inference chain is a problem.

--- DNA binding transcription factor activity 

  is_a (has_part?) DNA binding

if we have regulation of DNA binding transcription factor activity we don't want to say regulation of DNA binding (the same would be true of DNA binding transcription factor regulator activity).

  • This is true of all MF except for binding.
  • Has_part (or another relation) could work if the regulates inference chains stops over it

Identification of Pre-Existing Errors Through Imports

https://github.com/geneontology/go-ontology/pull/16419

GH project link

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

Minutes

  • On call: Barbara, Ben, David, Eric, Harold, Jim, Kimberly, Laurent-Philippe, Pascale, Peter, Tanya

Meetings

Montreal GOC Meeting

  • We mostly discussed the item regarding annotation review.
  • What should the priorities, scope be for this?
  • Are there additional resources that could be put on this, either curators or software?
  • What metrics can we use to show that the annotation review we're doing improves the resource?

Ontology Developers' workshop

  • We discussed travel arrangements - David emailed Lourdes about SOP.
  • Will discuss metabolic pathways (GO, Reactome, Rhea)
  • Signaling pathways are another beast; Ben is working on this as part of pathways2go
    • What versions of the pathways should be presented for curators to work from?
    • If/how complexes are represented in GO is one critical point for this work

Editors Discussion

Regulation inference chain

  • Checking in Protege, there is no 'regulates o has_part' -> regulates chain
  • We discussed the differences between different types of regulation of a molecular function
    • Localization as regulation
    • Regulating the activity when the gp is otherwise where it enables its function

Identification of Pre-Existing Errors Through Imports

https://github.com/geneontology/go-ontology/pull/16419

  • Importing a new version of PO found some GO errors in micropyle formation
    • Travis found unsatisfiable classes
    • These show up in Protege as terms with owl:Nothing
  • The logical definition referred to the plant structure, but all of the annotations (Drosophila), and the parentage in GO, refers to an eggshell
  • What are best practices for fixing things like this?
    • Start a new branch; fix GO
    • This needs a more detailed SOP in our documentation

Taxon Constraints

  • Proposal: keep a separate yaml file of taxon constraints
  • How would this new pipeline for taxon constraints work in practice?
Retrieved from "https://wiki.geneontology.org/index.php?title=Ontology_meeting_2018-10-01&oldid=71591"