Inconsistent IRIs Between eccode and rhea Sources

[jplfaria]

Description:

I have observed that the eccode and the rhea sources use different IRI schemes for referencing Uniprot resources:

• eccode example:
  Uses a Bioregistry URL, for example:
  https://bioregistry.io/uniprot:F4I907

• rhea example:
  Uses a PURL URL, for example:
  http://purl.uniprot.org/uniprot/F4I907

Is the difference in Uniprot IRI usage between eccode and rhea intentional? I realize this isn't primarily a ROBOT discussion, but while merging multiple ontologies (including eccode and rhea) using ROBOT, I've noticed that ROBOT merge relies on exact IRI matching. This discrepancy might lead to duplicate or misaligned entities in the merged ontology. I’d appreciate any clarification on the cause of the discrepancy.

[cthoyt]

This might be because of several recent improvements to automatic URI assignment. I guess you're referring to some specific files in obo-db-ingest?

[jplfaria]

Your guess is correct – I should have made that clearer. I am working with the following sources:

• rhea: https://w3id.org/biopragmatics/resources/rhea/rhea.owl.gz
• eccode: https://w3id.org/biopragmatics/resources/eccode/eccode.owl

In these files, I also observed a discrepancy regarding EC URIs. For example, in ECCODE:

    <!-- https://bioregistry.io/eccode:1.1.1.n10 -->
    <owl:Class rdf:about="https://bioregistry.io/eccode:1.1.1.n10">
        <rdfs:subClassOf>
            <owl:Restriction>
                <owl:property rdf:resource="http://purl.obolibrary.org/obo/IAO_0100001"/>
                <owl:someValuesFrom rdf:resource="https://bioregistry.io/eccode:1.1.1.301"/>

Whereas in Rhea:

    <!-- https://www.rhea-db.org/rhea/10016 -->
    <owl:Restriction>
        <owl:onProperty rdf:resource="http://purl.obolibrary.org/obo/RO_0002333"/>
        <owl:someValuesFrom rdf:resource="https://www.ebi.ac.uk/intenz/query?cmd=SearchEC&amp;ec=3.1.1.49"/>
    </owl:Restriction>

Notably, the latter IRI (from Rhea) does not resolve. I understand there have been recent discussions regarding EC ontology issues in the Bioregistry GitHub, and while I’m not sure if this discrepancy falls within that scope, I’m reporting my findings since I actively work with these ontologies on my science project.

I appreciate all the hard done in maintaining these resources, they have become crucial to my research.