Manager Call 2018-04-05: Difference between revisions

From GO Wiki
Jump to navigation Jump to search
mNo edit summary
 
(11 intermediate revisions by 3 users not shown)
Line 3: Line 3:
= Agenda =
= Agenda =
== NYC Meeting Agenda ==
== NYC Meeting Agenda ==
*Need to remind people to register & book rooms
*Need to remind people to register & book rooms: Rooms are still available
**Rate expires 5pm NY time this Friday (6 April)
*Start working on the agenda in more details
*Start working on the agenda in more details
**Add global ID prefix to all IDs (eg 'SGD' mandatory in SGD:S000000537) across all groups to explicitly declare source
**Add global ID prefix to all IDs (eg 'SGD' mandatory in SGD:S000000537) across all groups to explicitly declare source (need more info- this seems to be done the extent it will be)
**https://github.com/geneontology/go-ontology/milestone/5
**https://github.com/geneontology/go-ontology/milestone/5
**https://github.com/geneontology/go-annotation/milestone/2
**https://github.com/geneontology/go-annotation/milestone/2
Line 14: Line 15:
===Software===
===Software===
====GO website====  
====GO website====  
Chris:  
<strike>Chris: Seth is switching Drupal to administrative mode. This is to prevent the site being hacked due to recently uncovered security holes in drupal core. Note that there will be no updates to Drupal after this happens. Seth has warned about this in the past. We either need to commit resources to administering Drupal, or we need to switch the website out, e.g. for a static site built using github pages. My person preference is for github pages. But either way this will divert efforts from other tasks. </strike>
Seth is switching Drupal to administrative mode. This is to prevent the site being hacked due to recently uncovered security holes in drupal core. Note that there will be no updates to Drupal after this happens.
 
Seth has warned about this in the past. We either need to commit resources to administering Drupal, or we need to switch the website out, e.g. for a static site built using github pages. My person preference is for github pages. But either way this will divert efforts from other tasks.


Seth:  
Seth:  
Line 31: Line 29:
requires a less centralized role.
requires a less centralized role.


Discussion Summary:
No one person is in charge of maintaining GO site.  "People" (undefined) need to move site documents (could be done in a day), then make this new site pretty & usable (this will take much longer) before what is currently a smouldering danger spontaneously ignites and becomes inaccessible.
:;Action: All documentation needs to be transferred to GH, possibly in/near https://github.com/geneontology/go-site
*Seth will seed landing page, etc
*Website will be frozen for doc move
*Suzi L/Suzi A will work on quickly transferring documentation to be preserved in GH
**Predicted move during or after week of 16 April
*PG can help with content of doc pages- needs to know when site is frozen during move


:;Action: Longer term- Look at software to generate new site based on these markdown docs
*Timeline for this second step TBD (CM, SL, SC)


====Release pipeline====  
====Release pipeline====  
informing users/curators of the releases, so we can know what data we are looking at (Pascale: For annotation reviews this has been problematic for me)
informing users/curators of the releases, so we can know what data we are looking at (Pascale: For annotation reviews this has been problematic for me)
&rarr; This is already embedded in the site! However, it's currently broken... fix is in the works
*Plan on having 'Release Pipeline' presented on 24 April Annotation call
*Need better documentation of data flow posted on the wiki
**especially addressing GPAD derived from NOCTUA, file movement/modification between MODs/UP/GO Central
====New qualifier====
====New qualifier====
* Adding a qualifier to every annotation in a GAF file changes the file dramatically. It would impact any user or tool that filters annotations by looking for a null value in the qualifier field.  
* Adding a qualifier to every annotation in a GAF file changes the file dramatically. It would impact any user or tool that filters annotations by looking for a null value in the qualifier field.  
* Do we really want to make this large change to annotations in the gaf file, or do we want to encourage users who want the additional information to start consuming GPAD, where the qualifier is required?
* Do we really want to make this large change to annotations in the gaf file, or do we want to encourage users who want the additional information to start consuming GPAD, where the qualifier is required?
Addition of new qualifier was already discussed on an Annotation (possibly) call in March. 
Proposed implementation of 2.1 GAF (was May meeting) likely will be pushed back.
New solution: Take GPAD and GPI, have GO Central make the GAF (2.1) from that info.  This is the current pipeline at MGI. 
:;Action: JB to look into having all MODs link to a central GAF repository at GO Central.  Should alleviate differences in files b/w MOD and GO Central (more processing after files leave MODs).
:;Action: Documentation: work out relations (DH, PT, PG, KVB, CM)


== Working Groups/Projects ==
== Working Groups/Projects ==
Line 43: Line 70:
*[https://docs.google.com/document/d/1IEgyIHDMMZIMdFaR33jBaRg14d7-u6HKaI2couHTHXo/edit#heading=h.9nx0eh5kf2pz Draft Survey]
*[https://docs.google.com/document/d/1IEgyIHDMMZIMdFaR33jBaRg14d7-u6HKaI2couHTHXo/edit#heading=h.9nx0eh5kf2pz Draft Survey]
* Action from last meeting: (Kimberly): Move to Survey monkey, make changes discussed, and send it around
* Action from last meeting: (Kimberly): Move to Survey monkey, make changes discussed, and send it around
Survey will be sent out for last suggestions this week, will be released to each group & introduced on 10 April Annotation call.




'''''Time.  Rest of agenda bumped to next meeting'''''
=== Protein Complexes ===
=== Protein Complexes ===
=== HTP Codes ===
=== HTP Codes ===
Line 86: Line 116:
== GO Helpdesk ==
== GO Helpdesk ==
Discuss this ticket: https://github.com/geneontology/go-site/issues/584
Discuss this ticket: https://github.com/geneontology/go-site/issues/584
This is up for discussion at next Manager meeting.  Seems like an outreach task. There is some documentation on GO for curators- 1st step is to contact GO.  No guidelines on what GO should do (how to train, submit, who is responsible for maintaining...)
Email forwarded to Suzi A.


== Ontology  ==
== Ontology  ==
Line 96: Line 130:


= Minutes =  
= Minutes =  
*On call:
*On call: Pascale, Seth, Peter, Suzi L., David H, Judy, Kimberly, Chris, Suzi A.

Latest revision as of 13:03, 4 April 2018


Agenda

NYC Meeting Agenda

Working Groups/Projects

Software

GO website

Chris: Seth is switching Drupal to administrative mode. This is to prevent the site being hacked due to recently uncovered security holes in drupal core. Note that there will be no updates to Drupal after this happens. Seth has warned about this in the past. We either need to commit resources to administering Drupal, or we need to switch the website out, e.g. for a static site built using github pages. My person preference is for github pages. But either way this will divert efforts from other tasks.

Seth: To actually qualify this a bit more, now that I am looking at the issues (critical remote exploits) and possibilities to fix a bit closer, there is no "safe" mode available in our installation of Drupal. This means that we either migrate away very very fast indeed or I will drop it off the web temporarily today to do some of the needful updates. I believe the latter is the only realistic action as of right now.

Moving forward though, we should probably decide if there is to be a Drupal "role" or if we should hasten a migration to something that requires a less centralized role.

Discussion Summary:

No one person is in charge of maintaining GO site. "People" (undefined) need to move site documents (could be done in a day), then make this new site pretty & usable (this will take much longer) before what is currently a smouldering danger spontaneously ignites and becomes inaccessible.

Action
All documentation needs to be transferred to GH, possibly in/near https://github.com/geneontology/go-site
  • Seth will seed landing page, etc
  • Website will be frozen for doc move
  • Suzi L/Suzi A will work on quickly transferring documentation to be preserved in GH
    • Predicted move during or after week of 16 April
  • PG can help with content of doc pages- needs to know when site is frozen during move
Action
Longer term- Look at software to generate new site based on these markdown docs
  • Timeline for this second step TBD (CM, SL, SC)

Release pipeline

informing users/curators of the releases, so we can know what data we are looking at (Pascale: For annotation reviews this has been problematic for me) → This is already embedded in the site! However, it's currently broken... fix is in the works

  • Plan on having 'Release Pipeline' presented on 24 April Annotation call
  • Need better documentation of data flow posted on the wiki
    • especially addressing GPAD derived from NOCTUA, file movement/modification between MODs/UP/GO Central

New qualifier

  • Adding a qualifier to every annotation in a GAF file changes the file dramatically. It would impact any user or tool that filters annotations by looking for a null value in the qualifier field.
  • Do we really want to make this large change to annotations in the gaf file, or do we want to encourage users who want the additional information to start consuming GPAD, where the qualifier is required?


Addition of new qualifier was already discussed on an Annotation (possibly) call in March.

Proposed implementation of 2.1 GAF (was May meeting) likely will be pushed back.

New solution: Take GPAD and GPI, have GO Central make the GAF (2.1) from that info. This is the current pipeline at MGI.

Action
JB to look into having all MODs link to a central GAF repository at GO Central. Should alleviate differences in files b/w MOD and GO Central (more processing after files leave MODs).
Action
Documentation: work out relations (DH, PT, PG, KVB, CM)

Working Groups/Projects

GO Curation Survey

  • Draft Survey
  • Action from last meeting: (Kimberly): Move to Survey monkey, make changes discussed, and send it around

Survey will be sent out for last suggestions this week, will be released to each group & introduced on 10 April Annotation call.


Time. Rest of agenda bumped to next meeting

Protein Complexes

HTP Codes

RCA Evidence Code

  • Need to work with several groups to review their annotations:
    • Gramene (35815 annotations)
    • As an example, 2929 RCA Molecular Function annotations
      • Date: 20060831
      • Reference: GR_REF:8030
      • Manual review of mappings; can we suggest ISM?
    • GeneDB
      • Some annotations from reviews, but some to papers that describe a systems analysis that is probably more in line with the original intent of RCA
      • RCA -> CCA (combined computational analysis)??
    • EcoCyc
    • AgBase
    • AspGD
    • PAMGO_MGG
  • If we decide to keep RCA, then we'd need to remove the one-year limit; it doesn't make sense.

Documentation

QA/QC

Working Group Meetings

  • Quarterly?
    • Rotate between ontology and annotation (small, focused groups)
    • Test virtual meetings, but still focused on the task

Noctua

1.0 Rollout

  • Merge Tremayne's fork with Noctua-dev for further testing
  • Some outstanding issues on SAE
  • Single source file for MODs to consume

Reactome Imports

  • David, Kimberly have looked at some imports/translations
  • Ben's work looks to be potentially very useful and promising for seeding templates, models
  • Will probably require significant manual input, refinement, testing

End User Tools

GO Helpdesk

Discuss this ticket: https://github.com/geneontology/go-site/issues/584

This is up for discussion at next Manager meeting. Seems like an outreach task. There is some documentation on GO for curators- 1st step is to contact GO. No guidelines on what GO should do (how to train, submit, who is responsible for maintaining...)

Email forwarded to Suzi A.

Ontology

Announcing changes to the wider user community

  • This was brought up by Birgit in the protein complex merge ticket - she was suggesting we make an announcement. However we never made annoucements before (except for obsoletions); should we start? What would be the guidelines? For example when a merge affects more than 100, 1000 annotations ? EXP only ?
  • Pascale: changes to definitions of root terms: https://github.com/geneontology/go-ontology/issues/14899; isn't this a big change ? How should we notify people?
    • Should we create a label in GH 'Major change', so we can easily pull out the tickets and show them at Ontology and Annotation calls ? (perhaps in annual reports as well)
  • Pascale: suggestion: create a new label 'major change' and announce them either on twitter and/or on the website as user stories (or whatever it's called).


Minutes

  • On call: Pascale, Seth, Peter, Suzi L., David H, Judy, Kimberly, Chris, Suzi A.