GO Managers, Weds. April 9th, 2008 8 AM PDT, 9 AM MDT, 10 AM CDT, 11 AM EDT, 4 PM GMT
Present: Midori Harris, Pascale Gaudet, Judy Blake, Suzanna Lewis, Chris Mungall, Jane Lomax, Jennifer Deegan, David Hill.
Review Action Items from previous meeting
All: Check list of database contacts and make any additions needed. http://wiki.geneontology.org/index.php/External_Database_Contact_Info. Chris to email GO List. DONE
- ACTION ITEM: Mike/Chris need to agree on the date to put scripts in place. Put scripts in place. DONE
- Chris will send a pre-announcement to go-friends mailing list to announce the new file DONE
Reference Genome group: Decide on QC procedures, frequency, etc. [In the Ref genomes meeting agenda]
Reference Genome group: Determine protocol for making ISS annotations. [In the Ref genomes meeting agenda]
Reference Genome group: Try WebEx "raise hand" feature for next conf call. Pascale will put that in the agenda for ref genomes meeting [DONE - failed...]
Regulates loaded in AmiGO?
DONE - regulates fully implemented.
Removal of sensu endings and proper definition and naming of terms now finished.
Signaling overhaul started.
Initial ideas for structure at:
overhaul about to recommence
Lipoprotein terms for reactome and UCL
Various content items added to salt lake city agenda.
Page on wiki to collect all quality control mechanisms
Applying for ISSN for the newsletter..
advocacy/outreach topics for SLC:
Collaboration with Reactome to be discussed.
Jen: Emily has this in hand and is working with Esther Schmidt to make sure the necessary changes are made. The work is not yet ready but this is because Esther has been busy with other priorities.
Groups producing predictions
Does it make sense to bring in prediction annotations when they go stale within a year?
(Current policy is that we do not bring in predicted annotations. - Should this change?)
Things generally to consider:
- annotation assistance
- software needs
- web presence
Jane: We need to figure out how to prioritize.
Suzi's priority list:
- web presence
- reatome etc
- nurturing other new groups
- bringing in IEAs.
Web Presence group
Should Jane and Jen work with web presence to make release more co-ordinated and ensure that user needs are taken into account?
- General agreement that the web presence group needs more co-ordination. No clear plan on who should deliver that.
- Lack of biologists on the group means their needs are not represented.
- Web presence calls seem to be all about technical issues, so biologists who do join quickly drop out again. These calls should not be dominated by technical issues.
- Concern from Jen and Jane that any time they spend managing web presence is time that they do not spend on ontology content.
Conclusion: this group needs better co-ordination, but it's not clear who should provide that.
[Discussion between Jen and Jane after call:
Jane could perhaps spend half a day a week giving the biologist viewpoint and being release manager when needed.
She would not be technical manager though, as that could come from the software side.]
The cvs rearrangement is now all in place.
Encouraging people to use new file paths.
Question: Do we have a list beyond go-friends for our major users? e.g. uniprot etc.
Surely this is what go-friends is for?
Do we need an RSS feed?
Perhaps have something on download pages?
Quality control has been formalized:
Eliminating a bug that causes web search engines to hammer database.
Could have caught this with automated testing. Now improving this testing framework.
Comments on the wiki page welcomed software QC.
Software group want to collect in file system or on wiki list of derived files in cvs.
i.e. which files are derived from other files via a script that is run on a cron job, and who is responsible for those scripts?
Need to have a list of these on a wiki page:
New tracker system:
A demo is now available for people to try. [link?]
Seth sent a link for people to test AmiGO 1.6. Sent to web presence working group. Needs feedback.
Ben written bulk loading script to make loads happen quicker.
Demo of OBO-Edit plugin from Michael Schroeder's group. Michael is going to do some more polishing and send out a version to let people try. It is a text mining tool. We will set aside some time to demo it at the meeting. We will try to do it at a time where we can reach Michael if we have questions. Michael should come to Nomi or Chris to get this into the main source tree of OBO-Edit. David says a separate plugin may be best. So people can add it on if they want it. Will be very useful if it works well. We should have a go at it.
Is Goose safe against rogue SQL? Yes it can only bring down an internal database in Berkeley. It would not bring down AmiGO.
- Monthly call was yesterday; not too many people
- orthology determination (P-POD) is 'on schedule'
- curation tool for reference genomes is progressing
- Suzi, Judy, Michael made a nice 'annotation pipeline' document, available at: http://wiki.geneontology.org/index.php/Annotation_pipeline
- Annotators liked the document
- we'll meet next week to discuss it
- Pascale is having a hard time making people 'complete' the annotations; some people want to wait for the 'pipeline' to be in place, but she is not sure how different it will be
In MGI they have an internal wiki so they can co-ordinate who is doing which reference genome gene.
Maybe on wiki just post list of genes and people can mark if they are done. This is until the database is all ready to hold the ref genome stats. Database may be ready in a month.
Programmers working on annotation tool.
All on schedule.
BLAST part done.
Already preliminary version ready.
GO meeting agenda
Some discussion of how best to arrange the agenda.
Reference genome section should just be a report from ref genome meeting, so no lengthy discussion needed.
General point from Suzi:
We need to know who is responsible for what across the board.
Could add to agenda 'is there anything where we are missing a responsible person?'
We should have no duplication of effort, and no areas of work that are without a responsible person.
For all topics, managers should look at their individual areas on the agenda and adjust things to make sure the most importnat points are discussed in an appropriate order. Go the wiki and make adjustments.
Collaborations and Interactions
Operations and Procedures
Staffing and Personnel
- ref genome paper... Pascale would like to start working on that again