Ontology meeting 2013-07-11
- 1 EXISTS, TAGGED-INFERRED, NOT-ENTAILED
- 2 Plan for importing reactions from Rhea into GO
- 3 Follow-up: Hydrogen biosynthetic process
- 4 Follow-up: Giardia terms
- 5 Cycles in GO
- 6 Behaviour
- 7 New TG templates for organelle parts?
- 8 BUMPED FROM LAST MEETING (?): Organelle vs. macromolecular complex
- 9 BUMPED FROM LAST MEETING (?): Project Management in JIRA
- 10 BUMPED FROM LAST MEETING (?): MEMBRANE SIDES VS LEAFLETS
- 11 BUMPED FROM LAST MEETING (?): STRUCTURAL MOLECULE ACTIVITY
- 12 Molecule-level processes vs. population processes
- 13 Protein binding/ protein domain binding xps
EXISTS, TAGGED-INFERRED, NOT-ENTAILED
We need a strategy to deal with the growing list of these in the weekly assert report.
Suggestions so far: have the report tag new ones and old ones? It might be easier for us to go through and assert the ones we're sure we want to keep, and the ones we're still considering (recording the fact that we've done that somewhere) and ask Heiko to delete everything else?
Latest update: we decided we do need to look through the regulation links because they often indicate some underlying problem in the parallel 'regular' hierarchy.
TO DO: Jane will check cell cycle related ones before D/T go to JGI next Thurs and resolve all the remaining ones.
Heiko was going to try and figure out a way to only have the new EXISTS, TAGGED-INFERRED, NOT-ENTAILED links shown in the weekly report so that we can work through the original set systematically.
Plan for importing reactions from Rhea into GO
Let's discuss the plan, and figure out where we are wrt the xp pattern for these terms, and the xp set.
TO DO: Chris: need to figure out the directionality issue in OWL (bidirectional vs. unidirectional), come up with xp pattern to be used. One possibility: always have bidirectional parent, create both unidirectional children. For those cases where only one direction is physiologically reasonable, maybe only have that particular unidirectional term. Need to come up with good naming convention with the children terms, David has some ideas. TO DO: Jane will create JIRA issue/s to track this work.
CONCERN: What about existing annotations to the bidirectional term when the unidirectional kids are created? May be unsatisfactory to annotators who find a loss in specificity. Reannotation in fits and starts? This should not stop us.
As an addendum to this, David will give an update on an test case for a metabolic process.
TO DO: David will do pilot project with glycolysis, linking MF to BPs to describe the pathway. Probably not temporally just yet. Need to address reversible reactions. Create separate xp file with those relationships.
Follow-up: Hydrogen biosynthetic process
See background here http://wiki.geneontology.org/index.php/Ontology_meeting_2013-06-06#Follow-up:_metabolism_terms and Jane's email dated June 27th (subject Fwd: Re: Templates absent in Term Genie).
Suggestion: 'hydrogen generation' instead of 'hydrogen biosynthetic process'
TO DO: create 'hydrogen generation'(met process with an net output of H2) and manually place children underneath. Biosynthetic processes may be easy but metabolic processes will be harder. Need to look for net gain of H2 in the sum of reactions.
Follow-up: Giardia terms
Request from UC Davis group.
TO DO: Nail down the axis for orientation for left/right terms. We are leaning toward adding these terms. We will guide them in the use of existing terms that they can apply to their project. Jane will communicate with them.
Cycles in GO
I've spoken to Tony and it seems QuickGO only assumes GO is acyclic over is_a and part_of. Cycles over other relationships shouldn't cause a problem. We could try adding one to see what happens, but we would need to fix OE behaviour first.
TO DO: OE needs to be fixed to not throw the fatal error. Or turn off cycle check completely.
TO DO: Add check in Jenkins. Add this as a JIRA item.
There's a ticket which includes a proposal to update the def of behaviour . David has also suggested making a disjoint between behaviour and developmental process.
TO DO: Don't use the 'excluding development' in the definition but do make the disjoint relationship between behaviour and developmental process. David will do this.
New TG templates for organelle parts?
See this SF request: https://sourceforge.net/p/geneontology/ontology-requests/10242/
TO DO: Go ahead with template generation. Paola to create JIRA set of issues for this new template. Heiko says there are 5 templates in waiting that are in the queue already. Need to prioritize.
BUMPED FROM LAST MEETING (?): Organelle vs. macromolecular complex
Stemming from some observations on inferred links in the assembly node, e.g.:
(Added inferred link) GO:0000921 'septin ring assembly' GO:0070925 'organelle assembly'
GO:0000921 'septin ring assembly' also has the is_a parent 'cellular protein complex assembly'. This was added back in 2008 (http://www.ebi.ac.uk/QuickGO/GTerm?id=GO:0000921#term=history). However, 'septin ring' only has 'organelle' as ancestor, not 'protein complex'. Should we make 'macromolecular complex' and 'organelle' disjoint?
The proposal would be to have
organelle DisjointWith mm complex
'''NOTE:''' We discussed this at length at the last meeting and decided we wouldn't add this disjoint because some components are legitimately both organelles and protein complexes e.g. spindle
BUMPED FROM LAST MEETING (?): Project Management in JIRA
As we likely won't have time to go through this, may I suggest that we all look at our tickets and schedule progress on them if necessary.
Everyone look and update please.
BUMPED FROM LAST MEETING (?): MEMBRANE SIDES VS LEAFLETS
SF item from Alan Bridge: https://sourceforge.net/p/geneontology/ontology-requests/10223/
Is a membrane side equivalent to a membrane leaflet (one of the lipid layers)? If so, can all 'internal/external side of membrane' terms be changed to 'internal/external leaflet of membrane'?
- leaflet of membrane bilayer ; GO:0097478
- external side of plasma membrane ; GO:0009897
- internal side of plasma membrane ; GO:0009898
- GO:0031232 extrinsic to external side of plasma membrane
- GO:0031233 intrinsic to external side of plasma membrane
- GO:0031234 extrinsic to internal side of plasma membrane
- GO:0031235 intrinsic to internal side of plasma membrane
- GO:0031237 intrinsic to external side of plasma membrane, in periplasmic space
- GO:0031362 anchored to external side of plasma membrane
- GO:0071575 integral to external side of plasma membrane
TO DO: We agree side = leaflet. Becky: Keep the side in the primary term name and add leaflet as a synonym. Clarify terms to say that the side doesn't include the rest of the universe.
BUMPED FROM LAST MEETING (?): STRUCTURAL MOLECULE ACTIVITY
Is there a long-term plan for where we can't use annotation extension? E.g. for mappings. Looking for a term to map to KW: viral matrix protein, which is a structural constituent of virion.
TO DO: Add new term for the keyword mapping.
Molecule-level processes vs. population processes
How do we want to express these in logical definitions?
Nothing more at this time.
Protein binding/ protein domain binding xps
INFORMATIONAL: Chris did some work on logical xps for protein binding and protein domain binding. Need to talk about this some more next meeting.