search

information-artifact-ontology / ontology-metadata

OBO Metadata Ontology
Creative Commons Zero v1.0 Universal
19 stars 8 forks source link

OMO is not in the OWL EL profile #57

Open hlapp opened 4 years ago

hlapp commented 4 years ago

The reason is a few <owl:oneOf> axioms:

Use of ObjectOneOf with multiple individuals ObjectOneOf(<http://purl.obolibrary.org/obo/IAO_0000410> <http://purl.obolibrary.org/obo/IAO_0000420> <http://purl.obolibrary.org/obo/IAO_0000421>) [EquivalentClasses(<http://purl.obolibrary.org/obo/IAO_0000409> ObjectOneOf(<http://purl.obolibrary.org/obo/IAO_0000410> <http://purl.obolibrary.org/obo/IAO_0000420> <http://purl.obolibrary.org/obo/IAO_0000421>) ) in OntologyID(OntologyIRI(<http://purl.obolibrary.org/obo/omo.owl>) VersionIRI(<http://purl.obolibrary.org/obo/omo/2020-06-08/omo.owl>))]
Use of ObjectOneOf with multiple individuals ObjectOneOf(<http://purl.obolibrary.org/obo/IAO_0000002> <http://purl.obolibrary.org/obo/IAO_0000120> <http://purl.obolibrary.org/obo/IAO_0000121> <http://purl.obolibrary.org/obo/IAO_0000122> <http://purl.obolibrary.org/obo/IAO_0000123> <http://purl.obolibrary.org/obo/IAO_0000124> <http://purl.obolibrary.org/obo/IAO_0000125> <http://purl.obolibrary.org/obo/IAO_0000423> <http://purl.obolibrary.org/obo/IAO_0000428>) [EquivalentClasses(<http://purl.obolibrary.org/obo/IAO_0000078> ObjectOneOf(<http://purl.obolibrary.org/obo/IAO_0000002> <http://purl.obolibrary.org/obo/IAO_0000120> <http://purl.obolibrary.org/obo/IAO_0000121> <http://purl.obolibrary.org/obo/IAO_0000122> <http://purl.obolibrary.org/obo/IAO_0000123> <http://purl.obolibrary.org/obo/IAO_0000124> <http://purl.obolibrary.org/obo/IAO_0000125> <http://purl.obolibrary.org/obo/IAO_0000423> <http://purl.obolibrary.org/obo/IAO_0000428>) ) in OntologyID(OntologyIRI(<http://purl.obolibrary.org/obo/omo.owl>) VersionIRI(<http://purl.obolibrary.org/obo/omo/2020-06-08/omo.owl>))]
Use of ObjectOneOf with multiple individuals ObjectOneOf(<http://purl.obolibrary.org/obo/IAO_0000103> <http://purl.obolibrary.org/obo/IAO_0000226> <http://purl.obolibrary.org/obo/IAO_0000227> <http://purl.obolibrary.org/obo/IAO_0000228> <http://purl.obolibrary.org/obo/IAO_0000229>) [EquivalentClasses(<http://purl.obolibrary.org/obo/IAO_0000225> ObjectOneOf(<http://purl.obolibrary.org/obo/IAO_0000103> <http://purl.obolibrary.org/obo/IAO_0000226> <http://purl.obolibrary.org/obo/IAO_0000227> <http://purl.obolibrary.org/obo/IAO_0000228> <http://purl.obolibrary.org/obo/IAO_0000229>) ) in OntologyID(OntologyIRI(<http://purl.obolibrary.org/obo/omo.owl>) VersionIRI(<http://purl.obolibrary.org/obo/omo/2020-06-08/omo.owl>))]

I don't know whether meeting some OWL profile is a goal for OMO (presumably, at least OWL DL is, actually). I also know that most (all?) EL reasoners will simply ignore these axioms. It still seems odd to me that in essence this is trying to emulate a controlled vocabulary of values for a certain property, for which to me OWL axioms seem like a poor use.

Anyway, because of the above, at present any ontology importing OMO will by definition no longer be able to be in the EL profile.