GO-CAM Conference Call - 2019-01-15

From GO Wiki
Jump to: navigation, search

Meeting URL

  • Please consult the GO calendar

Agenda

  • From 2019-01-07 annotation conference call:
    • Import of whole genomes
      • Working meeting at Berkeley the end of this month
      • WormBase and MGI will be the first genome annotation sets to be imported
        • gp2term relations
        • evidence and annotation extensions
        • search, display, editing
        • gpad output
        • overall data flow
        • documentation
    • Continued discussion of modeling questions
      • Annotation of protein complexes and their respective members
      • Instance-level representation
        • When to annotate to the same 'instance' of a biological process vs different 'instances'?
        • Implications for curator workflow
      • Modeling from multiple species
        • Creating a model for one species when the evidence derives from experiments in multiple species
        • Creating models for processes that involve more than one species
    • Import of Reactome pathways
      • Glycolysis
      • MAPK
      • BMP
      • Wnt
      • ER-UPR
      • GPI synthesis
      • Size limitations and delineation of 'models'
        • How to 'stitch' or connect large models in a consistent manner that allows us to represent dependencies (upstream, downstream) in a biologically meaningful way that is also computationally amenable
    • Model manipulation
      • Generation and use of templates
      • Merging models
      • Copying and/or re-using annotations

Minutes

  • On call: Barbara, Chris, David, Dustin, Edith, Elena, Giulia, Harold, Helen, Kevin M, Kimberly, Laurent-Philippe, Marie-Claire, Paul T., Petra, Rob, Ruth, Sabrina, Seth, Stacia, Suzi A
  • Import of whole genomes
    • David, Kimberly, and the software team will have a f2f meeting to work on importing the first two whole genome's worth of annotation for WormBase and MGI
    • Will report back on a future call - decisions made, any problems that arose, etc.
    • For converting annotation extensions, Paul T. and Dustin have done some preliminary work which groups should look at on the corresponding wiki page:
    • Model manipulation will likely be key for dealing with whole genome's worth of annotation imports and we will need to gather explicit requirements for that
  • Continued discussion of modeling questions
  • Import of Reactome pathways
    • Smaller working group will be examining import of Reactome pathways into GO-CAMs
    • Specific pathways have been chosen based on preliminary work already done, e.g. glycolysis and signaling, ER-UPR
    • This work should, and already has been, very informative for how we model things in GO-CAM
    • Reactome imports also raises issues about model size and delineating and linking smaller models that are causally related
  • GO-CAM display
    • Petra has shown some of the dicty community GO-CAM models
    • Reaction was positive, but we need to spend more time working on various display options, making people aware of what APIs already exist for grabbing data for display, etc.
    • Whole genome import is currently the top priority, but intuitive, user friendly display of models is very important and will be the subject of future work