Update Noctua input / output -> primary input / output

[vanaukenk]

Project link

https://github.com/orgs/geneontology/projects/151

Project description

This project organizes the work needed to update Noctua to use primary input and output instead of input and output. Work is needed in: ShEx, annotations, UI, and possibly import code.

Should this be in scope:

Do we want only Noctua to be updated, or also our files/products, the visualization widgets and make sure contributing groups can load this data? Do we need to produce test files?
Technically we are not violating any formats, maybe this is OK.

Deliverables

TBD

PI

TBD

Product owner (PO)

Kimberly Van Auken

Technical lead (TL)

Seth Carbon

Other personnel (OP)

Jim Balhoff, Dustin Ebert, Pascale Gaudet, Tremayne Mushayahama

Technical specs

TBD (template: https://docs.google.com/document/d/111UqtS3G0aJZpAijZYI3Da0t94OQpGePlPJsqZE4Tio/edit)

Other comments

To do - GOC tasks (more specific actions will be recorded in tickets in appropriate repos):

• (DONE) Fix ontology to use has primary input/output as appropriate
• Add primary input/output relations to ShEx where needed - cross-check with ontology logical defs
• Remove transports_or_maintains_localization_of from ShEx
• Assess impacts on Reactome and yeast pathway imports - needs discussion
• Create TSV for updating existing relations in Noctua annotations/models
• Update Noctua UI: form data entry, form table, VPE
• Trial run of changes on noctua-dev
• Announcement of changes on annotation call
• Make changes during Noctua maintenance outage
• Update Visualization tools (GO-CAM Pathway viewer & all others)

Dependencies

• GOC pipeline: loading scripts for annotations must be fixed to recognize these new relations
• Upstream sources [pombase/GOA, others] need to be able to import and export
• Make sure anyone using the Pathway Widget has updated to the recent version (exact task to be defined better)
• Other applications that might use the 'has input' data for network analysis? Announce change more broadly.
• RO inference chains must be checked; probably OK because inherited from parent; confirm with Jim
• Update curator and user documentation

[pgaudet]

This is OK with me.

When do you plan to start this work? It would be useful to discuss this on a managers call to have everyone on the same page.