Node normalization issue in UI for Bivalirudin

NCATSTranslator / Feedback

A repo for tracking gaps in Translator data and finding ways to fill them.

7 stars 0 forks source link

Node normalization issue in UI for Bivalirudin #932

Open TranslatorIssueCreator opened 3 months ago

TranslatorIssueCreator commented 3 months ago

Type: Bug Report

URL: https://ui.test.transltr.io/

ARS PK: dd6d2da9-7a2d-4d8d-b5d1-42b33cd1ad99

Steps to reproduce:

what genes activity may be decrease bivalirudin?

Screenshots:

sandrine-muller-research commented 3 months ago

The first bivalirudin is ChEBI ID and the second is the PubChem ID Interestingly SRI NameRes service is grouping both ChEBI and PubChem in the same clique. Asking for @dnsmith124 's help.

dnsmith124 commented 2 months ago

Name Resolver returns both of these terms when searching for "Bivalirudin": https://name-lookup.test.transltr.io/lookup?string=bivalirudin&autocomplete=true&highlighting=false&offset=0&limit=10&biolink_type=SmallMolecule

Output: [{ "curie": "CHEBI:59173", "label": "Bivalirudin", ... }, { "curie": "PUBCHEM.COMPOUND:19797045", "label": "Bivalirudin", ... }]

@gaurav any insight here?

gaurav commented 2 months ago

Thanks for reporting this! At a minimum, that second clique should really be labeled "Bivalirudin trifluoacetate", but there's actually a third "Bivalirudin (trifluoacetate)" clique that is a molecular mixture. I've opened https://github.com/TranslatorSRI/Babel/issues/346 to figure this out fully, but I think I should be able to just combine these three cliques manually for Hammerhead.