Problem:
The current Biolink Model has a ChemicalAffectsGeneAssociation, but doesn't have a GeneAffectsChemicalAssociation. Some KPs have a significant number of edges of the latter format, but is unsuccessfully attempting to annotate those edges with qualifiers (and qualifier value ranges) only defined in the former association class. This leads validation errors (as of reasoner-validator 3.8.8 validating against Biolink Model 3.5.4).
Example:
This legitimate edge does not validate:
subject: TRPC4
predicate: affects
object: Barium
object_aspect_qualifier: transport
object_direction_qualifier: increased
What working group (or team) did this request originate from?
Molepro KP
Describe the solution you'd like
Add GeneAffectsChemicalAssociation class annotated with suitable qualifiers and qualifier value ranges/subproperties. A simple search for the inverse association won't necessarily work since the inverse predicate is affected_by not affects.
Tag relevant members for discussion
@sierra-moxon @mbrush @vdancik
Problem: The current Biolink Model has a
ChemicalAffectsGeneAssociation
, but doesn't have aGeneAffectsChemicalAssociation
. Some KPs have a significant number of edges of the latter format, but is unsuccessfully attempting to annotate those edges with qualifiers (and qualifier value ranges) only defined in the former association class. This leads validation errors (as of reasoner-validator 3.8.8 validating against Biolink Model 3.5.4).Example:
This legitimate edge does not validate:
subject: TRPC4 predicate: affects object: Barium object_aspect_qualifier: transport object_direction_qualifier: increased
What working group (or team) did this request originate from? Molepro KP
Describe the solution you'd like
Add
GeneAffectsChemicalAssociation
class annotated with suitable qualifiers and qualifier value ranges/subproperties. A simple search for the inverse association won't necessarily work since the inverse predicate isaffected_by
notaffects
.Tag relevant members for discussion @sierra-moxon @mbrush @vdancik