Ontology meeting 2015-10-29: Difference between revisions

From GO Wiki
Jump to navigation Jump to search
No edit summary
 
(13 intermediate revisions by 4 users not shown)
Line 1: Line 1:
Attendees:
Attendees: DavidOS, Paola, Melanie, Chris, Heiko, Harold, Tanya, PaulT


Minutes:
Minutes: DavidOS


=== Discussion of GO annotation tracker ===
  Apparently no one from the GOA team is looking at this.  GOA curators not aware of it.
  Chris: But it was moved from Sourceforge, with multiple announcements to GO discuss?
  David: Where should ontology editors send reports on effects of edits on annotations?
  Melanie: We should send things to GOA help. 
  But this doesn't => tracking/discussion.  And does it go to all annotators?
  Decision - add to this tracker and then send an email to GOA help.  Anything else?
  Lots of things on this tracker are UniProt keyword-related issues.
  Chris: why do these (UniProt keywords) still exist?  Can't they be replaced by a subset of GO terms?
  AI: Chris to bring up on Managers call.


=== Let's move Jira tickets to GitHub ===
=== Let's move Jira tickets to GitHub ===
Line 8: Line 20:
Background: http://wiki.geneontology.org/index.php/Ontology_meeting_2015-09-17#Review_of_Jira_tickets
Background: http://wiki.geneontology.org/index.php/Ontology_meeting_2015-09-17#Review_of_Jira_tickets


Given that an automated solution is not yet in sight, Chris advises that "we should do this manually and ruthlessly". So let's all go ahead and move our own tickets to GH. (Paola will look at tickets assigned to non-editors and follow up with relevant people.)
Given that an automated solution is not yet in sight, Chris advises that "we should do this manually and ruthlessly". So let's all go ahead and move our own tickets to GH. (Paola looked at tickets assigned to non-editors, and re-assigned to ticket creator or followed up (with Seth).)


In the process, please close anything that is now solved or out of date, move anything that belongs in other GH trackers, etc.
In the process, please close anything that is now solved or out of date, move anything that belongs in other GH trackers, etc.
   
 
  All agree.
  Decision: pattern requests go to design_pattern tracker.  
  Editors should write design patterns on repo and get them validated (DOS, CM and HD working on automating validation)
  If to be used for TG template, make linked ticket on TG tracker once pattern is finalised.
 
==='cell proliferation' terms===
 
Most have logical def, e.g. 'epithelial cell proliferation' is_a 'cell proliferation' that acts_on_population_of 'epithelial cell'. Did we have plans for a TG template? (Not in Jira.)
 
Note: this is a case where we have a semi-random distribution of terms in the ontology vs annotation extensions.  Should we just make terms for all AE usage and retrofit AEs?
 
  Should we have a template?
  Yes. Paola will request it.
  Many are in AEs - should all those go into the ontology?
  Melanie:  Isn't this going to cause bloat?  Shouldn't we be pushing in the other direction?
  David: Doing this would hide the annotation from term enrichment (only inferred parent classes would be visible)
  Paola: Enrichment to such specific terms would be rare.
  Decision: once the template is ready, create terms for children of 'cell proliferation' that have abundant annotation extensions.
  Then look at the structure under 'cell proliferation': if it's too flat, see if we can create some intermediate terms (using the template).
  Paola can do it (not as a top priority though).
 
===Adding logical defs to cellular response terms===
DOS has started to do this. What's the syntax, and should we look into filling all gaps.
 
PUNTED TO NEXT WEEK
 
=== Update on status and plan for Synapse-related GO terms ===
 
Timeline given to Broad showed first-pass ontology terms released in November
 
  David:  Synapse CC terms mostly done. 
  Synaptic process - getting close to a draft refactoring in Google Docs.
  May be able to get most added in November.
  No pointing in adding until this is (mostly) done.
  Progress slow because the cycle between VU proposal and GO counter-proposal is slow.
  Trying to speed this up.  May visit Amsterdam again.
  Broad Project Manager (Rosy) is aware of status and process.  Is trying to help speed up.
 
=== Follow-up on Inference management ===
Documentation of ETINE flagging & relationship to redundancy stripping.
  PUNTED TO NEXT WEEK
 
=== Follow-up on Design Patterns ===
 
Next steps
  PUNTED TO NEXT WEEK
 


[[Category:Ontology]]
[[Category:Ontology]]
[[Category:Meetings]]
[[Category:Meetings]]

Latest revision as of 11:29, 30 October 2015

Attendees: DavidOS, Paola, Melanie, Chris, Heiko, Harold, Tanya, PaulT

Minutes: DavidOS

Discussion of GO annotation tracker

 Apparently no one from the GOA team is looking at this.  GOA curators not aware of it.
 Chris: But it was moved from Sourceforge, with multiple announcements to GO discuss?
 David: Where should ontology editors send reports on effects of edits on annotations?
 Melanie: We should send things to GOA help.  
 But this doesn't => tracking/discussion.  And does it go to all annotators?
 Decision - add to this tracker and then send an email to GOA help.  Anything else?
 Lots of things on this tracker are UniProt keyword-related issues.
 Chris: why do these (UniProt keywords) still exist?  Can't they be replaced by a subset of GO terms?
 AI: Chris to bring up on Managers call.

Let's move Jira tickets to GitHub

Background: http://wiki.geneontology.org/index.php/Ontology_meeting_2015-09-17#Review_of_Jira_tickets

Given that an automated solution is not yet in sight, Chris advises that "we should do this manually and ruthlessly". So let's all go ahead and move our own tickets to GH. (Paola looked at tickets assigned to non-editors, and re-assigned to ticket creator or followed up (with Seth).)

In the process, please close anything that is now solved or out of date, move anything that belongs in other GH trackers, etc.

 All agree. 
 Decision: pattern requests go to design_pattern tracker.  
 Editors should write design patterns on repo and get them validated (DOS, CM and HD working on automating validation)
 If to be used for TG template, make linked ticket on TG tracker once pattern is finalised.

'cell proliferation' terms

Most have logical def, e.g. 'epithelial cell proliferation' is_a 'cell proliferation' that acts_on_population_of 'epithelial cell'. Did we have plans for a TG template? (Not in Jira.)

Note: this is a case where we have a semi-random distribution of terms in the ontology vs annotation extensions. Should we just make terms for all AE usage and retrofit AEs?

 Should we have a template?
 Yes. Paola will request it.
 Many are in AEs - should all those go into the ontology?
 Melanie:  Isn't this going to cause bloat?  Shouldn't we be pushing in the other direction?
 David: Doing this would hide the annotation from term enrichment (only inferred parent classes would be visible)
 Paola: Enrichment to such specific terms would be rare.
 Decision: once the template is ready, create terms for children of 'cell proliferation' that have abundant annotation extensions. 
 Then look at the structure under 'cell proliferation': if it's too flat, see if we can create some intermediate terms (using the template).
 Paola can do it (not as a top priority though).

Adding logical defs to cellular response terms

DOS has started to do this. What's the syntax, and should we look into filling all gaps.

PUNTED TO NEXT WEEK

Update on status and plan for Synapse-related GO terms

Timeline given to Broad showed first-pass ontology terms released in November

 David:  Synapse CC terms mostly done.  
 Synaptic process - getting close to a draft refactoring in Google Docs.
 May be able to get most added in November.
 No pointing in adding until this is (mostly) done.
 Progress slow because the cycle between VU proposal and GO counter-proposal is slow.
 Trying to speed this up.  May visit Amsterdam again.
 Broad Project Manager (Rosy) is aware of status and process.  Is trying to help speed up.

Follow-up on Inference management

Documentation of ETINE flagging & relationship to redundancy stripping.

 PUNTED TO NEXT WEEK

Follow-up on Design Patterns

Next steps

 PUNTED TO NEXT WEEK