Links to PRO entities - Githubissues

geneontology / noctua-visual-pathway-editor

0 stars 0 forks source link

Links to PRO entities #81

Open ValWood opened 6 months ago

ValWood commented 6 months ago

Links go to https://proconsortium.org/cgi-bin/pro/entry_pro?id=PR:000050512

should go to https://proconsortium.org/cgi-bin/entry_pro?id=PR_000050512

pgaudet commented 6 months ago

Weird we should be using PURLs - this is what's in the dbxref file: entity_types:

type_name: protein type_id: PR:000000001 id_syntax: ([OPQ][0-9][A-Z0-9]{3}[0-9]|[A-NR-Z]0-9{1,2}){1}(-[0-9]+){0,1}|[0-9]{9} urlsyntax: https://purl.obolibrary.org/obo/PR[example_id] example_id: PR:Q64445 example_url: https://purl.obolibrary.org/obo/PR_000060258

@kltm is this what is implemented? Are the PURLs not resolved correctly? (this make it a PRO issue, not GO).

kltm commented 6 months ago

It would be good to see the links in context--I'm not sure where this is occurring. Usually, this kind of thing would be driven by the dbxrefs.yaml in go-site, but there may be an issue in information propagation or hard-coding.

ValWood commented 6 months ago

The xref file specifies PR_000060258 Noctua is using PR:000060258

kltm commented 6 months ago

@ValWood Can you give a link to where in Noctua you are seeing this?

ValWood commented 6 months ago

In my model meiotic cohesion protection at anaphase I Double-click on entity rec8. You will see the PR identifier under rec8 Spom in the right hand panel

kltm commented 6 months ago

Links: http://noctua.geneontology.org/editor/graph/gomodel:66187e4700001744 http://noctua.geneontology.org/workbench/noctua-visual-pathway-editor/?model_id=gomodel:66187e4700001744

Replicate:

Go to http://noctua.geneontology.org/workbench/noctua-visual-pathway-editor/?model_id=gomodel:66187e4700001744
Scroll to "bottom" of model
hover over "rec8 Spom" box
click on [pencil] in halo
hover over "PR:000050512" link
http://www.proconsortium.org/cgi-bin/pro/entry_pro?id=PR:000050512 is destination

The expectation would be https://purl.obolibrary.org/obo/PR_000060258 (go-site metadata), or at least "correct-ish" https://proconsortium.org/cgi-bin/entry_pro?id=PR_000050512 .

kltm commented 6 months ago

Going to https://amigo.geneontology.org/amigo/gene_product/PomBase:SPBC29A10.14 , then adding direct term filter "topological DNA co-entrapment activity", we get an annotation with isoform PR:000050512, that links to https://purl.obolibrary.org/obo/PR_000050512 . This would seem to indicate that the metadata, at least some places, is healthy.

kltm commented 6 months ago

Noting that "noctua-visual-pathway-editor/src/@noctua.curie/data/go-context.ts" seems correct. My next best guess would be package.json in noctua-visual-pathway-editor:

    "amigo2": "^2.4.3"

...unfortunately, I'm having trouble building the local packages for the moment (I may need to ask @pkalita-lbl for help here). It may be worthwhile first to clarify with @tmushayahama where this metadata is coming from: an npm package, go-site, or the contexts file?

kltm commented 6 months ago

I think we need @tmushayahama 's input here before acting. @tmushayahama , can you pinpoint where that information is coming from, in case packages need a refresh or if it's wired in some other way?

vanaukenk commented 5 months ago

Noting that the incorrect link is also happening for PRO entries on the Noctua form editor.

pgaudet commented 4 months ago

ping @tmushayahama , can you pinpoint where that information is coming from, in case packages need a refresh or if it's wired in some other way?