Ontology meeting 2015-08-27
Attendees: DavidOS, DavidH, Chris, Heiko, Melanie, Paola, Heiko, Harold
Minutes: Tanya
DC meeting planning
- Noctua will be good for editors and curators! Starts with and ends with of processes will be hugely beneficial.
Introduction and explanation will be done to bring new annotators up to speed and ready to use this new tool.
- Annotation extensions : we will show the new 'inferred' GAF that Jenkins generates that folds
the AEs so that they can be 'sucked up' and used by existing tools, particularly QuickGO (TonyS needs to know where they are so he can make use of them), clarify that this 'experimental' inference file will now become production
- MF refactoring:DOS will finish editing and will send out MF refactoring document later today to eds for final edits before sharing more widely before meeting
- breakout with PaulT at the DC meeting to discuss the plan and the specifics of implementation
- protein complex: review of links between GO, IntAct, PRO, scopes, definitions, equivalence classes, intersections, benefits.
- How can we reduce redundant effort / mine each other's resources to make our work easier?
- Potential role for PRO in classifying by component, based on mining IntAct
- How can we limit GO scope to be manageable while giving other resources, such as IntAct, what they need?
- Extreme case: get rid of all GO complex terms apart from those defined entirely by function/location. But (how?) would this support dynamic classification for IntAct?
- Middle ground: Only allow terms that can be completely axiomatised - this could potentially include composition (e.g. by protein family => generic SMAD complex)
- Less extreme: limit classification by component to homo-heterodimersing protein families where possible (a role for PRO in this?)
- How can we reduce redundant effort / mine each other's resources to make our work easier?
- Standardising protein production/secretion classes:
- TG FF discouraged terms include secretion; regulation of X secretion
- Critique of use of X secretion classes
- some arguments here seem to apply even more strongly to production terms, for which Pascale has written a critique.
- Defense of X production terms
- Deepening annotations - develop ontology patterns e.g. protein binding
Provisional plan to do move all protein binding + annotation extnsion to protein ID. Posthoc classification can come from leveraging Panther (which needs to be OWL-ised). Sketch here: https://docs.google.com/document/d/1uevfMJ5FhBnEY_g3_Xh1ZQBPucBRtO0C5pnQvj7qWZ0/edit#heading=h.cj82byymmujx
Why is ligand-gated ion channel activity not under receptor activity ? - BUMPED
These are generally referred to as receptors by biologists.
See https://github.com/geneontology/go-ontology/issues/12029
Jira tickets -BUMPED AGAIN
Time for our periodic review of Jira tickets, especially unassigned ones. Please go through JIRA issues in advance of meeting. We will aim to devote about 20' of the call to discuss Jira tickets.