Ontology meeting 2013-04-25

From GO Wiki
Jump to: navigation, search

Minutes: Tanya

Attendees: Jane, David, Chris, Heiko, Harold, Tanya, Judy, Paola

Project Management in JIRA

Jane to sort out/close website related JIRA issues.
Abandon idea to add Interpro to add structure to the protein domain binding branch.  Make our own.
Paola:Some progress made on met/bio/cat xps but response from CheBI is slower.

Follow-up: cell-type templates

See http://wiki.geneontology.org/index.php/Ontology_meeting_2013-04-18#Cell_type_templates

All children of GO:0097285 cell-type specific apoptotic process now map to a CL ID.

Chris/Heiko to report on feasibility/timeframe for TG template.

 Paola and Chris edited CL and x-cell.owl to add new terms and xps.  Host of problems/issues 
arose which need to be addressed first having to do with word usage.  Will be tackled. CL has 
cell types and anatomical cell types.  Need reasonable grouping classes like 'neuron apoptotic process.'
 Paola: when template is done, create grouping subclasses for apoptotic processes 
like 'neuron apoptotic process.' 

"Think of Christmas trees.  Things hanging at reasonable looking levels. Not a 
bazillion ornaments and only a few branches. All of the baubles were at the top of 
the tree, instead of spread down to the lower branches." Jane and Chris talking with 
someone at a EBI/industry workshop.
 after CL apoptosis, Pascale requests CL differentiation.   Chris will put x-
cell.owl into gene_ontology_xp_write.obo sometime before Tuesday. Look at new 
links report - PAOLA - please double check. Template will follow in the next couple of weeks.
 After May 9th, PO xps will go next with integration to xp file and template roll out

Follow-up on SF upgrade

See http://wiki.geneontology.org/index.php/Ontology_meeting_2013-04-18#SF_upgrade

ACTION ITEM: Jane: upgrade SF 


This discussion is about content of the ontology-related pages. Can we somehow link to some of the Drupal pages that are finished like the ontology download page? It seems that the whole website is going to take some time.

 Can make a link on the existing page that directs to the beta page. ACTION: Chris will do.

Subset metadata

Tony is adding the 'do not annotate' anti-slims to the DB so they can be used by in QuickGO and protein2go. They'd prefer to have the subset comments at the level of the subset itself rather than as a comment added to each individual term that belongs to that subset. Is there any way we had do this in OBO format, in the header or something?

Jane will speak to Tony about convention. May add terse description and larger description to 
OBO header as metadata for now.

Cell cycle issues

starts_with ends_with

Is there any reason we can't start using these in xps in gene_ontology_write.obo? I need them for the cell cycle logical defs.

ACTION: Chris: sort out the existing relationships and place them in the ontology. In two weeks, we can use.
ACTION: Jane: to Gofriends: Warn for two weeks that new relationships (starts_with, ends_with) are 
coming to gene_ontology_write.obo, then roll them out.  Add link to new downloads page on beta site.
ACTION: Harold will check what file MGI loads.  Will let Jane know.


For between phases and transitions. I need this before I can progress editing. Is this the correct relation? What's the process for adding a new relation to gene_ontology_write.obo?

see above

InTact xrefs

Birgit from InTact has been adding lots of complexes to cc via TG FF. They would like to have reciprocal links back to GO for InTact complexes. These will be the species-specific entities, one for each species so a max of about 8 per GO term. Updates to the xrefs could be handled manually at this stage. Two questions for us: are we happy to have the xrefs? Would we consider adding a general dbxref field in TG FF so Birgit could add the xrefs as she goes along?

ACTION: Harold to interact directly with Birgit.  Heiko to add dbxref field for Intact to TermGenie.

Pending obsoletion subset?

I've been thinking it might be a good idea to have a 'pending obsoletion' subset for marking terms we intend to obsolete. This would allow us to easily identify terms we'd slated for obsoletion, and also would allow browsers to display this info to users.

Subset vs. comment, which one to use.  Subset has benefit of being able to filter and 
color.  OR use a standard string in the comment and filter on the standard string.  REAL ACTION: 
use comment field and add in the reason for obsoletion.

Protege 4.3

Chris: Recommend that everyone install this.  If you use code generation plugin, version 4.2 erase your home directory.
 A Protege 4.3 release build is now available on our website.

> Link for registered users:
> http://protege.stanford.edu/download/registered.html#p4.3
> New users, please register before downloading:
> http://protege.stanford.edu/download/register.html
> Release Notes
> ---------------------
> - Fixed a bug in the code generator that in some cases caused deletion of a user's home directory.  
Java code generation is accessible from Tools | Generate Protege-OWL Java Code.