Managers 16Aug06: Difference between revisions
Jump to navigation
Jump to search
Line 120: | Line 120: | ||
::Judy will plan for 25 | ::Judy will plan for 25 | ||
*Action item: advisory board meeting presenters to prepare talk | *Action item: advisory board meeting presenters to prepare talk drafts in time for Aug 30 call | ||
===AOB=== | ===AOB=== |
Revision as of 08:39, 29 August 2006
GO Managers call 16 August 2006
Calling in: Michael Ashburner, Judy Blake, Rex Chisholm, Jennifer Clark, Jane Lomax, Midori Harris, David Hill, Eurie Hong, Suzi Lewis, Chris Mungall
Minutes: Midori/Rex
some preliminary chat about traveling to Hinxton ...
David - plans for is_a complete paths in biological process
- timing - some during MGED9 in Seattle
- actual paths
- need 'developmental process' as is_a parent (and sib of development)
David does agenda on the fly ...
Action items from last call
- Jane will document is_a complete CC
- not up yet, but ready to go
- Jim Hu's GO wiki - David to send address
- done
- split annotation camp into computational vs. manual
- based on experience with latest camp; they were duly warned that emphasis was on literature curation
- mainly for Mike C/SGD
- So will there be a 'computational-annotation' camp?
- Judy - Mike C did say something about organizing it, maybe Sept/Oct; idea is we would promote SOPs
- Will we use RCA for InterPro2go, TMHMM, etc?
- wanted to try, but MGI ran into some difficulties
- annotation camp recommendations for GOC
- many at camp thought the policies were being set then & there
- campers did think some should go to GOC
- need input from Michelle Gwinn on proposed RCA usage
- many at camp thought the policies were being set then & there
- Eurie to start next newsletter
- well under way; Kimberly will send it out
- includes all items on list from previous call
- well under way; Kimberly will send it out
Reference Genome progress
- Rex has sent first gene list
- too time-consuming for Rex to provide orthologs; MODs etc. to look up; seems to be working so far
- will send UniProt IDs where available, otherwise Entrez Gene IDs
- protein seq most useful for orthology determination
- also consider connection between OMIM and UniProt
- David - please use HGNC-approved standard name
- will send UniProt IDs where available, otherwise Entrez Gene IDs
- another problem - in some cases, there's an enormous amount of literature; may have to refine understanding of 'complete annotation'
- does 'complete annotation' for now mean using a set of recent + key refs, rather than all available?
- David - risk of missing original basic biology
- mouse has only 10-20 papers (cf S. pombe 80)
- these genes very well studied in other species
- Rex - issue is how to balance; different studies, different genes, different organisms
- intentionally picked fundamental biology for first set, so most or all spp have orthologs; maybe alternate with genes where, say, mouse has lots of the experimentation
- David - maybe put whole list somewhere for dbs to tackle at own pace (with tracker), rather than distribute piecemeal? MGI will have to pick & choose papers for some genes
- Rex envisions people annotating, and annotations flowing into CVS by normal route; will assume a week's genes are done unless he hears otherwise (but note WB flow, with curation and testing on 4-6 week cycle)
- hard to know when tasks are accomplished; want to minimize differenCes from dbs' usual procedures
- David - tracker would only need to address Ref Genomes list; chart with genes and each db's progress; put on wiki
- Judy - someone can fill in UniProt IDs & HGNC names; db reps can fill in their IDs
- Rex - ask for ID and percent of literature annotated per gene
- David - risk of missing original basic biology
- does 'complete annotation' for now mean using a set of recent + key refs, rather than all available?
- Action item: Rex (with help as needed) to put spreadsheet on wiki as table, with new columns as requested for UniProt, HGNC, etc. Run draft by GO managers; notify Reference Genomes list when it's ready.
- Rex welcomes input on points from longer ref genome email
- Rex will continue sending emails to both reference-genome and GO managers
User Advocacy progress
- Newsletter
- Group priorities on wiki
Annotation Outreach progress
- Waiting to hear from PIs on priority list
Ontology Development progress
- CNS development additions committed
- progress on immunology (Alex)
- PAMGO content meeting - about 150 new terms (Jane, Amelia, PAMGO cuRators)
- New tracker wish list (David & Midori)
- Action item: Midori to put wish list on wiki
- Many people want quick way to get new terms in
- David - allow people to submit OBO stanza
- Suzi - need faster initial response
- Suzi - three issues: ease of submission; ranking ease of implementation; response turnaround time
- Jen - gets faster response phoning (or Skype) than SF comments or email
- aside on Skype
- David - we have a list of MOD curators who are ontology development group members; don't hesitate to ask them about requests in their area
- Jen - gets faster response phoning (or Skype) than SF comments or email
- Suzi - may be able to do better than SourceForge now
Software group
- AmiGO progress
- Seth visited SGD to gather requirements, etc.
- want to do something more 'with-it'; go in new directions with AmiGO
- AmiGO working group IRC tomorrow? (there was one yesterday)
- Suzi & Chris to join next AmiGO chat, whenever it is
- Eurie - waiting for Ben Hitz to return from vacation
- TAIR switched to RefSeq IDs for sequences, which has slowed loading (4-fold!!!); updates no longer 3x/week at the moment
- some discussion of technical aspects and possible fixes; SGD programmers are looking into it
- TAIR switched to RefSeq IDs for sequences, which has slowed loading (4-fold!!!); updates no longer 3x/week at the moment
- considering using Chado as underlying db, with bridge layer to be interoperable with existing schema
- a few technical points; stay in touch with Gail Binkley (SGD DBA) as this progresses
- John's history toolkit proposal
- is the managers list the best place to discuss that sort of thing?
- several people say yes
- also good to include software people for technical aspects
- trial would be feasible, using obodiff; use as basis for future development
Advisory Board Meeting
- Any comments on the agenda?
- Judy - working with Peter Tarczy-Harnach (advisor); he's found a room
- PIs will send out CD to participants, with material from StX meeting, research portion of grant, study section reviews and our response, etc.
- Suzi - basic time frame settled, e.g. presentations from 8 AM - 1 PM
- please add comments to wiki; also any other material to put on CD
- Judy - could try to get presentation drafts beforehand - in time for next managers call
- David & Midori to tag-team on biological aspects of content
- Judy - logistics; 20 people listed; suggestions for additional people (e.g. Monte)
- Craig Neville-Manning can't come this time
- Judy will plan for 25
- Action item: advisory board meeting presenters to prepare talk drafts in time for Aug 30 call
AOB
- Rex - table on wiki (for people who joined late; see above)
- wiki table setup is ugly; may not work for 20-25 columns -- and hundreds of genes -- we'd need; is there any other technology we could use?
- try Google spreadsheets
- Ontology stuff from Swiss-Prot meeting (Suzi)
- round table of people interested in disease ontology; want to collaborate but also have to get on with own projects
- Rex - we've put in tons of work so far
- have an outline (tho not one that Barry would like)
- has multiple parentage
- may be good enough to start with, tho horribly flawed
- will fix problems, but have to ignore much of what Barry & Anand recommend
- Suzi - we now have a window; prospective users can't wait
- Rex - many problems easily fixable, e.g. incorrect parents, incomplete paths
- Owen White, Win Hide and others can try to use it, and give feedback
- Rex - file is 25 MB
- put on ftp site?
- reason not on public web yet is that it includes cross-refs to SNOMED
- Suzi - might be fine; only has xref, nothing more, and xref can be ignored
- Michael - would SNOMED object to the xrefs?
- Rex - that's the worry
- Michael - if you use only the SNOMED ID, there's no proprietary info ...
- Rex - there's also a version With the SNOMED xrefs stripped out
- Michael - Yves Lussier is on SNOMED board; Rex has talked to him; MA will 'put pressure' on Yves
- make it available one way or another, then make sure others (e.g. Owen, Win) can map to their systems
- latest isn't on OBO site yet (v3beta30)
- organized using FMA and GO biological process
- have an outline (tho not one that Barry would like)
- Action item: Rex to make file available to GO managers by Aug. 23rd
Next call
August 25, 8 AM PDT, 10 AM CDT, 11 AM EDT, 4 PM BST
agenda: Midori
minutes: Jen
(from the rota)