Pattern error on CHEBI pesticide role

[pbuttigieg]

@kaiiam I think this is one of your design patterns.

pesticide is a CHEBI role, so the DP axioms don't hold. Picked this up with HermiT.

Not sure how we'd like to fix that - we could assert:

'concentration of'
 and ('characteristic of' some 
    ('chemical entity' and 'has role' some pesticide
     and ('part of' some 'surface water')))

[kaiiam]

@kaiiam I think this is one of your design patterns.

I don't remember doing this. It looks like it is based on the original concentration DOSDP which Chris originally setup but modified to have characteristic in the axiom. Strange stuff.

'concentration of' and ('characteristic of' some ('chemical entity' and 'has role' some pesticide and ('part of' some 'surface water')))

This axiom is saying that it's a concentration and has role some pesticide. Not sure if qualities can have roles following RO.

[kaiiam]

has_role def

a relation between an independent continuant (the bearer) and a role, in which the role specifically depends on the bearer for its existence

So a quality (specifically dependent continuant) can't have a role.

[cmungall]

This axiom is saying that it's a concentration and has role some pesticide

it's easy to misinterpret nested expressions, but this isn't in fact what it's saying. The has-role applies to the chemical entity which is what we want (with various caveats about role modeling in chebi).

I'd rather avoid nesting altogether but nesting chebi roles into material entities in this way is not uncommon (GO names these expressions)

[kaiiam]

Thanks @cmungall for clarifying.