Manager Call 2019-03-06
- Agenda: Huaiyu
- Minutes: Suzi
- Update on all ongoing projects: https://github.com/orgs/geneontology/projects -
- Are goals clear for the GOC meeting next month?
- Is anyone encountering any roadblocks?
- We need % effort from the different people involved in project for all projects with a delivery date/active projects
- Created new projects for (1) GPAD/GPI specs (2) Documentation (to list items we have worked on during the current period)
- 'Gantt-chart' to manage projects: Hackathon tasks: how is that working ?
- The issues with the new input and output relations have been resolved. It is fine to make an official release.
GO meeting agenda
- Moving existing annotations to GO-CAM
From last meeting minutes (by far not exhaustive)
- Create working group to coordinate integration of IEAs (single source ideally): priority?
- Handling redundant information: priority?
Fate of "unmanaged" external2go files
New IEAs (not InterPro2GO)
We decided earlier to not accept IEAs from outside groups, as all sequences would be run through the dependable InterPro2GO after uploading- but this group/method *seems* like they have more accurate annotations and higher coverage.
Long discussion last week - What's the next step ?
We were going to set up a call - has it be scheduled?
Update on the GO-CAM model copy (Laurent-Philippe). Further discuss user requirements (e.g. contributors and template aspect)
Examples (temporary URLs):
- original DBL-1 pathway | duplicated DBL-1 pathway
- original Signaling by BMP | duplicated Signaling by BMP
Update on projects:
Documentation: this is just an ongoing list
- Pipeline [Seth]
- data issues, tend to be important for users
- PAINT pipeline: duplicated annotations for very few genes, some with slightly different withs, in last two releases. Cause likely evidence tracking issue. Dustin is working on this.
- Some genes missing from PANTHER upstream, Dustin/Seth will get to this.
- External2GO- remnant of domain switching.
- AmiGO is slowly being fully automated,
From Alliance call:
- Push GAF file to Alliance (right now for gene descriptions, probably other uses in future)
- [Judy] GO data put into a format that can be semi-automated, and in balance with what they're doing
- [Seth] Pushing data (annotation files) instead of pulling data
- Mapping human to HGNC in GO Central release
- [Paul] this has been the plan for a while, but hasn't been put in place
- [Pascale] How to get to the Uniprot entry page, may be more informative
- [PT] Consistent ID namespace whether they're protein coding or not
- [Judy] Alliance uses HGNC, so this would be consistent
- [MC] when to link to gene/protein, eg UniProt is explicitly protein
- [PT] Start with human, but also Xenopus and others need to have IDs clarified
- [Kimberly] Also need to address links in with/from throughout pipeline
- [PT] ...or of submission requirements
- need plan to act on- scope? all IDs or just with from, who we're working with, is this an internal exercise, etc.
[David] GPI file currently solves some of this, ties all entries to a GP
- GPAD-GPI specs
- headers- this will be addressed. Needs more attention in specs
- SO vs MSO [Judy] on agenda for Alliance all hands call in April
- Moving existing annots to GO-CAM
- [David] manually reviewing annotations to make rules
- [David] sometimes easier to make some changes in bulk import instead of one-by-one in Protein2GO, proposed a "special" GPAD for the initial load
- [Kimberly] in contact with LP and Tremayne
- [PT] time to bring newest (in-progress) version to dev for testing before release- Tremayne and Seth to work on this.
- [LP] model copying feature is ready for testing, URLs in agenda
- input as to requirements welcome
- [PT] future uses of copy function- all except EC and references, other uses, templates, etc.
- [David] Have add individual prompt templates?
- [Seth] very difficult now- keep this as requested feature for now
- [LP] also copies usable for teaching, no repos yet
- present this at GO meeting
Kimberly/David/Peter/Huaiyu in talks with Reactome, lots of details getting ironed out
QA rules: Eric hopefully getting to this, IBA getting filtered out
Website: few tickets remain, documentation and email for helpdesk remain
Other/off agenda: [David] release can be started today including PURLS pointing back at current. Everything defaults to release.
Prioritize projects/features because a user asks or shows a need?
- Eg term-ancestor table from https://github.com/geneontology/helpdesk/issues/165
- yes, this case- this will be needed for Alliance, internally, etc.
Pascale, David, Huaiyu, Judy, Laurent-Philippe, Suzi, Mike, Seth, Kimberly, Paul T