Manager Call 2020-07-29

From GO Wiki
Jump to: navigation, search

Agenda

  • Agenda: Kimberly [Pascale off]
  • Minutes: Laurent-Philippe
  • Present:

GO Baltimore

Where are we ? There is no page for it at the moment on http://wiki.geneontology.org/index.php/Consortium_Meetings_and_Workshops Do we organize a go user meeting ?

BCC2020 was organized with a new tool with virtual representation of a conference: https://remo.co/ Could ease breakout and functionality for interactive posters but each breakout limited to 8 people and the solution is not free.

Action item: Laurent-Philippe will contact Michelle to get an update and see if she had something plan for the GO user meeting

GO-CAM Jamboree

  • Proposal to have a GO-CAM curation jamboree sometime this fall
  • How critical will it be to have Noctua on a more secure production footing by this time?

Not sure yet before or after GO meeting. Possible concern with the fire season.

Name: GO-CAMboree ?

GPAD/GPI

  • There is a discussion about possibly duplicating effort by asking groups to submit a GPI file to GO and a BGI file to the Alliance.
    • Stacia raised the issue of possibly overlapping goals here on a GO annotation call some time ago.
    • Should we re-visit this and discuss aligning these files more closely? Could one file be derived from the other?
    • Should these types of files be moving towards a more ontological representation of entities? We'd ultimately like to do that with annotatable entities in GO.
    • What would the development costs be for everyone to make changes?
    • Need to communicate our thoughts/decision on this to the Alliance PIs
  • Would like to formalize groups’ review of the specifications and ask groups to sign off their review on a github ticket.
  • Would give groups until September 1st, but after that no changes would be allowed for 2.0.

Also how to include xrefs (eg gp2protein, gp2rna) in GPI files.

Data Visualization Workshop at ECCB

Are we interested in trying to register as a group?

https://bionetvisa.github.io

GO Wiki Mediawiki installation

Last week, a hiccup on the go wiki machine again brought up considerations for long-term maintenance. When we get to finally taking this over from Stanford, we'll need to do one of 1) migrate to a hosted mediawiki service; 2) allocate standing resources for instance and software maintenance, updates, backups, and recoveries; 3) migrate away from mediawiki (e.g. RTDs, GH wiki).

Ticket, will require some time and maintenance. Look at other solutions ? https://docs.readthedocs.io/en/stable/

Projects update

https://docs.google.com/document/d/1URLkEQBewlzD-NVAjbPLWD5Gu-5xq0RysTrm1GMbuiM/edit#


Legacy data

Laurent-Philippe

Discussed with Stuart about the archive; got a new version easier to work with in local.

Started the script to remap SVN to something that would fit the structure of release.geneontology.org

Quick test on 6 random dates to recreate releases from SVN: https://geneontology-tmp.s3.amazonaws.com/index.html#releases/

Some data that we may want to keep or reuse somewhere: gp2protein, gp2rna, tutorials, file format (eg gpad 1 etc), ontology diffs, enrichment gene sets (genes -> expected terms)

GH repository to keep track of development & share comments/issues: https://github.com/geneontology/archive-reconstruction

Moving groups to Noctua (#3)

Kimberly has a GPAD file with full history and looking into loading this David is working on GPAD/GPI 2.0 files

GO pipeline solidified / Infrastructure / DevOps (#4)

Discussion: Seth: nothing to report Had no time to work on the pipeline Pascale asked if we could pay for any services to make the pipeline more efficient. Seth: Nothing on the main pipeline, but the Travis checks (go-ontology) perhaps. We need to check

Seth: nothing to report Paul: taxon constraints ? Seth: fix in progress for allowing more commands in Jenkins. Will have to circle back to improve the general process

Example: most annotations of E.Coli to 1665, but other strains annotated and sometime to the species itself. Strategy to discuss for taxon ids. Uniprot and MODs have different taxon ids strategy. Eg SGD one taxon for all genes; Uniprot taxon assigned per gene. How to have it consistent in GO ? If no progress on discussion, we could have another step in the pipeline to remap to our taxon space for consistency.

GO-CAM content

i. List of BP to annotate ii. Annotation jamboree(s)

-> Do we want to schedule that ? Or are there Noctua features required first ? - Kimberly generated a list of pathways for C elegans - Kimberly: We *could* schedule a jamboree for the fall, but she is concerned that we have many questions about how to model certain things (namely, activity-regulating processes) - Kimberly will ask at the next annotation call to see who would be interested in attending, and will schedule a date

GO-CAMs from existing pathway resources

Ben is writing the paper


Creating GMT files

https://github.com/geneontology/go-site/issues/1536 Does this needs to be added in the priorities ?