search

FlyBase / flybase-controlled-vocabulary

The home of the FlyBase miscellaneous ontologies - incorporating the Drosophila Phenotype Ontology (see also FlyBase/drosophila-phenotype-ontology)
Creative Commons Attribution 4.0 International
9 stars 5 forks source link

Loss of non-redundant axiom in fbcv.obo #76

Closed Clare72 closed 5 years ago

Clare72 commented 5 years ago

We are losing this relationship from FBcv:0000525 'chemical' in fbcv.obo and fbcv.owl:

is_a: FBcv:0000452 ! origin of mutation

This results in chemical terms falling under owl:Thing rather than 'origin of mutation' in the fbcv hierarchy. This also affects fbcv-full, but not other files.

In fbcv-edit.obo:

[Term] id: FBcv:0000525 name: chemical namespace: origin_of_mutation def: "$sub_CHEBI:24431" [CHEBI:24431] synonym: "mutation induced by chemical exposure" EXACT [] is_a: FBcv:0000452 ! origin of mutation

In fbcv.obo:

[Term] id: FBcv:0000525 name: chemical namespace: origin_of_mutation def: "A chemical entity is a physical entity of interest in chemistry including molecular entities, parts thereof, and chemical substances." [CHEBI:24431] synonym: "mutation induced by chemical exposure" EXACT [] is_a: CHEBI:24431 ! chemical entity

In both files chemical entity has:

equivalent_to: FBcv:0000525 ! chemical

matentzn commented 5 years ago

This unboxed a scary can of worms.. Hahha. I think I found a workaround, but we this will get rid of redundancy stripping..

dosumis commented 5 years ago

Sounds familiar. Thought we'd fixed this a couple of months back. We can live with redundancy in the full release.

matentzn commented 5 years ago

This case we solved was a problem with OORT. But this here is something really weird. It is caused merely by running reduce (no filter/remove stuff of any kind).