Annotation Conf. Call 2021-11-16

Agenda and Minutes

Annotation Issues

Generic GO BP Slim Updates

• Update from Pascale
• https://github.com/geneontology/go-ontology/issues/22294

• We discussed implementation of the new generic GO BP slim. A few points:
  • There may be cases where both a parent and child term are represented in the slim.
  • In those cases, is it clear to users what the numbers for each term mean, e.g. in QuickGO the number assigned to the parent term is the remainder of what otherwise was not assigned to the child term.
  • One suggestion was to have a slim-specific label for the parent term like 'other DNA metabolic process' to make it clearer to users that this is used for annotations to the other child terms of DNA metabolic process
  • Integrating analyses with the matrix tool could be helpful for annotation QC
  • AI: Put together requirements for what we want a GO slim tool to do

Just to clarify. In QuickGO if there is a parent - child relationship then the annotation is only associated with the child term not the parent, see: https://www.ebi.ac.uk/QuickGO/annotations?goUsage=slim&goUsageRelationships=is_a,part_of,occurs_in&goId=GO:0006631,GO:0006629&taxonId=9606&taxonUsage=descendants&geneProductSubset=Swiss-Prot&geneProductId=UniProtKB:A1L0T0&geneProductType=protein

In QuickGO if a term is a child of 2 parent terms that are in the SLIM then the annotation is associated with both terms, see https://www.ebi.ac.uk/QuickGO/annotations?goUsage=slim&goUsageRelationships=is_a,part_of,occurs_in&goId=GO:0006631,GO:0006575,GO:0006399,GO:1901135,GO:0071941,GO:0006629,GO:0016073,GO:0006790,GO:0016071,GO:0005975,GO:0044281&taxonId=9606&taxonUsage=descendants&geneProductSubset=Swiss-Prot&geneProductId=UniProtKB:A0A024RBG1&geneProductType=protein

Annotation Extension Relations

Further Discussion of 'has input' and 'has output'

• Capturing metal ions as inputs and outputs
• GO classes using 'has input' or 'has output' in extensions
  • Metalloenzymes
  • Metal-dependent binding
  • Sensor activity
  • Channels/transporters/transport
  • Other BP
    • Response to....
    • Homeostasis
    • Chemotaxis

AmiGO: Change default closure to is and a_part_of and remove regulates

https://github.com/geneontology/amigo/issues/620 Justification is that this would align AmiGO behavior with Panther and QuickGO; more consistent for users

Attendance

• On call: Antonia, Bob, Chris, David, Dmitry, Dustin, Edith, Giulia, Harold, Helen, Kimberly, Li, Malcolm, Midori, Pascale, Patrick, Petra, Raymond, Rob, Ruth, Seth, Suzi, Tanya