Managers 10Oct07

GO Managers, Weds. Oct. 10 8 AM PDT / 10 AM CDT / 11 AM EDT / 4 PM BST

Agenda/chair: Midori
Minutes: Pascale
Present: Chris, Midori, Jen, Pascale, Judy, David

Review Action Items from previous meeting

• Display replaced_by & consider tags in AmiGO (Chris/swug)

DONE, Chris: you can see this in AmiGO already although the way it's done needs to be cleaned

• Finish & submit NAR update (Midori)

DONE, accepted

• Several meeting preparation tasks - reports, wiki additions, etc. (all managers)

DONE, obviously

Individual Manager reports

Ontology Development

• Jen and David are cleaning the muscle and cardio file and will send out an announcement within the next week or so
• Lung development is next: there will be a contents meeting about this

Staffing and Personnel

John has been Googled!!

Publications

• NAR manuscript accepted!
• Current Protocols in Bioinformatics article update (Judy & Midori)

Other Current Topics and Concerns

• Specific items (with relevant management groups):

Annotation: Do we want to encourage submission of annotations of alternate spliceforms?

Judy summaries: the ref genome group is aware that the annotation of multiple splice forms can pose problems: - we need to know which product we are annotating and we need to know which gene this belongs to

- Do we need another column?

- Chris: at the RefG meeting we talked about using a mapping, similar to the gp2protein file, which can have a one-to-many relationship

- Chris: can I have a feel for how many groups do it? MGI (structured notes), rat, who?

- Suzi: would everyone NEED to to it?

- Judy, no

- Chris: we can make a proposal: two ways of doing it: for eg, if MGI has two splice variants, they use the gene ID, and if they know the splice variant, they include that in the structured note; the other way is like UniProt, which maps all back to the 'parent' gene. UniProt also uses s suffix (people dont really like to encode information in the ID)

- two options: add splice variant ID in gene_id column of in new column

- Suzi prefers the first option because it seems 'cleaner'; annotate to the actual product having activity; however several issues with that (), so generally option 2 is favored

- Suzi: does that also include non-gene things like ESTs?

-David: that rises an interesting point: if there is an unpublished variant

[ACTION ITEM]: Chris will write a proposal for how to deal with this (including adding a new column?)

Microarray annotations

- Chris: we do not allow more granular annotation?

- Pascale: isn't that dealt with in column 12 or so? and use SO terms there

- Yes but everyone has 'gene' there

- Should we not be annotating to gene products? yes, but most people do not do it yet

- Midori: we hadn't decided which part of SO we can use

-Suzi: I thought all SO terms were allowed (that made sense in that context)

- Suzi: back to microRNAs: there is the gene, the transcript, and the gene product (processed transcript)

- David: how does the SO treat the miRNAs?

- All gene products are grouped by functional categories; for mRNAs, is the SO talking about the pre-miRNA or the processed miRNA?

- Chris: I think they are both represented. If the SO needs to be fixed we'll do it.

- David: also, the miRNA field is changing really fast [ACTION ITEM]: Think about updating documentation to state that all terms from SO are accepted. Chris: we need to do this with the splice variants.