LEGO February 1, 2016

From GO Wiki
Jump to: navigation, search

https://bluejeans.com/969313231

Agenda

  • Software updates
  • GAF updates
  • Relations documentation
    • Review draft?
  • Plans for upcoming meetings
    • ISB Meeting, Geneva, April 10-14
    • GO Meeting, Geneva, April 15-16
  • Models discussion
    • C. elegans insulin signaling - http://noctua.berkeleybop.org/editor/graph/gomodel:5667fdd400000696
      • Technical issues:
        • Adding annotations for genes not yet in NEO
        • Slow autocomplete for entities, GO terms, and evidence codes, especially
        • Visualization options for different evidence codes, i.e. places where the evidence is strong vs underwhelming
        • Add both evidence and comment with one click
      • Modeling questions:
        • For a signaling pathway, if/how to represent the BP without cluttering up the display
        • Example of a cross-species interaction experiment as evidence for ligand-receptor binding
        • Redundancy in has_input and directly_activates?
        • Use of the directly_activates relation and the question of being part_of a pathway or regulating a pathway
        • Using less granular MF terms, e.g. protein kinase activity vs protein serine/threonine kinase activity

Minutes

  • On call: Chris, David OS, Giulia, Heiko, Helen, Kimberly, Pascale, Ruth, Seth, Stacia, Suzi
  • Software updates
    • Noctua front page now takes curators directly to model search/selection
    • Delete model will be made a state option
    • Only models marked production will be loaded into AmiGO that is visible to the public
    • Large model layout issue fixed
    • IDs added to term names
  • GAF
    • David H, Heiko, Kimberly working on this
    • Need to decide how to handle regulates relations, e.g. directly_activates, when the MFs are really part_of a process
    • Files for MODs to consume will be on Jenkins
    • GPAD coming - need to sort out multivalue attribution
  • Documentation
    • Google doc is still editable
    • Relations documentation needs comments from Paul T., David OS
  • GO Meeting - Geneva, April 2016
    • Run another Noctua workshop for GO consortium curators?
  • Annotation issues
    • Need to be able to annotate to non-NEO (Noctua Entity Ontology) genes/gene products
      • Currently, minerva will not validate these entities
    • Autocomplete is sometimes slow
      • SOLR issue?
    • Visual display of different evidences
      • Some evidences provide stronger support than others
      • Explore ideas of how to display evidence differently, e.g., bold, more colors, etc.
    • Annotation comments
      • May be desirable to specify type of comment, e.g. supporting evidence sentence vs curator comment
    • Insulin model discussion
      • Adding part_of relations to a model can make it very cluttered - display options here?
      • Cross-species annotations - is there a way to provide a visual cue in the display?
      • has_input - is this relation redundant when the relation to the next MF is directly_activates?
        • What is the logical definition of a ligand and receptor?
          • e.g., insulin-activated receptor has_part insulin binding activity?
    • Curators should think about what types of questions could be asked of LEGO models