GO-CAM November 22, 2017

Zoom URL

https://stanford.zoom.us/j/679970729

Agenda

Updates/Discussion on 1.0 Release Milestones

Annotation Attribution in GPAD Output Files

• noctua github ticket #458 - Attribution with GO-CAM exports to GOC annotation files
• noctua github ticket #502 - providedBy should be used in GPAD export
• go-site github ticket #453 - All Noctua curators should have "updated" metadata associated in the YAML file (including group and URI)
• Issue summary: we want to make sure that, in the GPAD files, attribution in the 'Assigned By' field is correctly populated with a curator's annotation group
• Is this a good time to clean house wrt older or test models?
  • Step 1: Curators need to have entries in the groups field of the users.yaml file
    • Some groups still need to review their users.yaml (i.e. they are missing groups and also sometimes even missing orcids):
      • USC
      • UniProt (SIB and EBI)
      • PomBase
      • FlyBase
      • SynGO
      • Xenbase
    • Groups that are okay:
      • BHF-UCL
      • dictyBase
      • MGI
      • SGD
      • TAIR
      • WB
      • Zfin
  • Step 2: Convert any remaining GOC:nn identifiers to orcids
  • Step 3: Determine if, for any models, curators would not want their default groups to be used to populate providedBy value
  • Step 3: Populate providedBy value with appropriate group for each evidence statement 'Contributedby' a given curator
• Question: What is the significance of the model-level providedBy and how is it populated?

Display user and group labels from minerva metadata

• noctua github ticket #380 - Display user and group labels from minerva metadata

Allow negation in Noctua and SAE and in GPAD export

• noctua github ticket #524 - Allow negation in Noctua and SAE and in GPAD export

export-lego-to-gpad-sparql should collate by MOD

• go-site github ticket #431 - export-lego-to-gpad-sparql should collate by MOD

MInutes

• On call: Barbara, Chris, Edith, Harold, Jim, Kevin, Kimberly, Pascale, Paul T., Sabrina, Seth, Suzi

Agenda

Updates/Discussion on 1.0 Release Milestones

Annotation Attribution in GPAD Output Files

• AI: Seth/Berkeley will implement a Jenkins job that checks, for all entries in users.yaml that have edit permission in Noctua,
  • there exists an orcid
  • there exists at least one group (that is also present in groups.yaml)
• AI: Entries in users.yaml file that don't meet these requirements will be output in the Jenkins report and users will be asked to update

• AI: Berkeley will generate a models report that lists, for each model, :
  • model name
  • all curator names associated with that model
  • default group for each curator
• AI: Curators will be asked to review this report and if there are models for which their default group is NOT what they want in the providedBy field, then they need to indicate the appropriate group

Display user and group labels from minerva metadata

• Changed from 1.0 Milestone to wishlist as this impacts the contributor display on the landing page and is not essential for curation or GPAD output

Allow negation in Noctua and SAE and in GPAD export

• We reviewed the 1.0 milestones in the SAE github tracker
  • Restrict selectable evidence codes to manual subset
  • Bare MF annotation should generate part-of triples
  • The SAE should be aware of user groups and allow users to select and use them
  • Warn (prevent?) users if the model contains edges that could potentially be lost
  • Add a NOT button to each aspect
• AI: Berkeley will report back when these are implemented and/or when they need curators for further testing of the SAE
• Note that a newer version of the SAE will allow adding extensions like inputs to MFs (i.e. has_input(entity)), but causal relations will still be added in the graphical UI

export-lego-to-gpad-sparql should collate by MOD

• Will be implemented imminently