gouttegd
commented
1 year ago As I noted in the original ticket, the problem is that identifiers.org returns an invalid link for FMA, and the link extractor based on identifiers.org is tried before the OBO Foundry link extractor (which would return a correct link).
This is primarily an issue for identifiers.org, but it does raise some questions for Protégé:
- Should we continue using a identifiers.org-based link extractor?
- Should we replace it with a bioregistry.io-based one?
- Should the list of extractors (at least the ones that are based on external services such as identifiers.org) be configurable by the user, so that a user could for example decide that a bioregistry.io-based extractor should be used before the identifiers.org-based one?
Originally posted at https://github.com/obophenotype/uberon/issues/2923.
Clicking FMA hyperlinks in Protégé direct to URL of the format https://bioportal.bioontology.org/virtual/2006/FMA:##### with the message "The page you are looking for wasn't found. Please try again."
However, clicking the FMA links in OLS will direct to the FMA class in OLS. The PURL for the FMA class resolves to an Ontobee page, as expected.
For example, UBERON:0002079 'left cardiac atrium' has dbxref annotation FMA:7097.
In Protégé, clicking the FMA hyperlink directs to page https://bioportal.bioontology.org/virtual/2006/FMA:7097 with message "The page you are looking for wasn't found. Please try again." In OLS, clicking the dbxref link with the FMA directs here: https://www.ebi.ac.uk/ols4/ontologies/fma/classes/http%253A%252F%252Fpurl.obolibrary.org%252Fobo%252FFMA_7097?lang=en In OLS, the FMA PURL is http://purl.obolibrary.org/obo/FMA_7097 and resolves to: https://ontobee.org/ontology/FMA?iri=http://purl.obolibrary.org/obo/FMA_7097 Should Uberon FMA links in Protégé resolve to Ontobee or some other valid page instead of a page that states "The page you are looking for wasn't found. Please try again."?