cardiac muscle cell (sensu Arthopoda) unsatisfiability

[balhoff]

The Ubergraph unsatisfiability check reported a new problem. @gouttegd @bvarner-ebi does this look like anything that recently changed in CL?

cardiac muscle cell (sensu Arthopoda) SubClassOf Nothing

• cardiac muscle cell (sensu Arthopoda) EquivalentTo cardiac muscle cell and (develops from some cardioblast (sensu Arthropoda))
  • cardiac muscle cell SubClassOf develops from some cardiac muscle myoblast
    • cardiac muscle myoblast EquivalentTo myoblast and (part of some cardiac muscle tissue) and (develops into some cardiac muscle cell)
    • cardiac muscle tissue SubClassOf part of some heart
      • heart SubClassOf only in taxon some Vertebrata
      • Vertebrata SubClassOf Craniata
        • Craniata SubClassOf Chordata
        • Chordata SubClassOf Deuterostomia
      • only in taxon SubPropertyOf: in taxon
    • part of SubPropertyOf: overlaps
  • develops from SubPropertyOf: has developmental contribution from
  • cardioblast (sensu Arthropoda) SubClassOf only in taxon some Arthropoda
    • Arthropoda SubClassOf Panarthropoda
    • Panarthropoda SubClassOf Ecdysozoa
      • Ecdysozoa SubClassOf Protostomia
• overlaps o in taxon SubPropertyOf: in taxon
• in taxon some Protostomia DisjointWith in taxon some Deuterostomia
• has developmental contribution from o in taxon SubPropertyOf: in taxon

Axiom Impact

Axioms used 1 times

• cardiac muscle cell (sensu Arthopoda) EquivalentTo cardiac muscle cell and (develops from some cardioblast (sensu Arthropoda)) [cl-base.owl]
• cardiac muscle myoblast EquivalentTo myoblast and (part of some cardiac muscle tissue) and (develops into some cardiac muscle cell) [cl-base.owl]
• cardioblast (sensu Arthropoda) SubClassOf only in taxon some Arthropoda [cl-base.owl]
• cardiac muscle cell SubClassOf develops from some cardiac muscle myoblast [cl-base.owl]
• Ecdysozoa SubClassOf Protostomia [taxslim.owl]
• Arthropoda SubClassOf Panarthropoda [taxslim.owl]
• Chordata SubClassOf Deuterostomia [taxslim.owl]
• Vertebrata SubClassOf Craniata [taxslim.owl]
• Panarthropoda SubClassOf Ecdysozoa [taxslim.owl]
• Craniata SubClassOf Chordata [taxslim.owl]
• heart SubClassOf only in taxon some Vertebrata [uberon-base.owl]
• cardiac muscle tissue SubClassOf part of some heart [uberon-base.owl]
• in taxon some Protostomia DisjointWith in taxon some Deuterostomia [taxslim-disjoint-over-in-taxon.owl]
• part of SubPropertyOf: overlaps [ro-base.owl]
• only in taxon SubPropertyOf: in taxon [ro-base.owl]
• develops from SubPropertyOf: has developmental contribution from [uberon-base.owl]
• overlaps o in taxon SubPropertyOf: in taxon [uberon-base.owl]
• has developmental contribution from o in taxon SubPropertyOf: in taxon [ro-base.owl]

Ontologies used:

• uberon-base.owl (http://purl.obolibrary.org/obo/uberon/uberon-base.owl)
• cl-base.owl (http://purl.obolibrary.org/obo/cl/cl-base.owl)
• taxslim.owl (http://purl.obolibrary.org/obo/ncbitaxon/subsets/taxslim.owl)
• taxslim-disjoint-over-in-taxon.owl (http://purl.obolibrary.org/obo/ncbitaxon/subsets/taxslim-disjoint-over-in-taxon.owl)
• ro-base.owl (http://purl.obolibrary.org/obo/ro/ro-base.owl)

[gouttegd]

I (and others) have been adding some taxon restriction in several places, but I don’t recall adding one anywhere near the classes involved here…

I think the problematic bit is this one:

• cardiac muscle tissue SubClassOf part of some heart

That tissue should be part of circulatory organ, which is taxon-neutral, instead of heart, which is vertebrate-specific. (Though this will require some careful checking: cardiac muscle tissue is a striated muscle tissue – are we sure that all heart-like organs are always made of striated muscle?)

In any case it does not look like this axiom was introduced recently, so I am concerned that the issue has never been flagged before. I did a complete merge of Uberon+CL+FBbt (plus the corresponding bridges) two days ago and didn’t catch anything.

[balhoff]

@gouttegd this check builds and integrates various ontologies from their source, with the aim of catching problems before release: https://github.com/INCATools/ubergraph/blob/master/ci/test-merged.sh — so it may be slightly different content from what you merged.

[gouttegd]

Thanks! That’s the kind of cross-ontology CI checks that I would like to see broadly adopted!

And right, I know why I didn’t catch anything: because I am stupid and I had not included the taxslim-disjoint-over-in-taxon.owl component in my own check, meaning that I was not set up to catch any taxon-related problem… 🤦

And now I have ~3,600 larval interneurons that are somehow unsatisfiable. :scream:

[balhoff]

@gouttegd thanks so much! The check is green again.