search

NCATSTranslator / ReasonerAPI

NCATS Biomedical Translator Reasoners Standard API
35 stars 28 forks source link

Is the possible data type (scope) of TRAPI Edge attribute_type_id property wider than biolink:association_slot? #453

Closed RichardBruskiewich closed 1 year ago

RichardBruskiewich commented 1 year ago

Here's the definition of the Edge attribute_type_id property:

        attribute_type_id:
          $ref: '#/components/schemas/CURIE'
          description: >-
            The 'key' of the attribute object, holding a CURIE of an ontology
            property defining the attribute (preferably the CURIE of a
            Biolink association slot). This property captures the relationship
            asserted to hold between the value of the attribute, and the node
            or edge from  which it hangs. For example, that a value of
            '0.000153' represents a p-value supporting an edge, or that
            a value of 'ChEMBL' represents the original source of the knowledge
            expressed in the edge.

The question here is concerning the scope of this field. If not a biolink:association_slot

  1. Could it be a Biolink category class (e.g. biolink:OrganismTaxon)? see below
  2. Could it be a Biolink predicate slot term (i.e. a reification of an edge to another concept value)?
  3. Are there any other legitimate values that it could have?
RichardBruskiewich commented 1 year ago

On item #1 above, I think I have a fresh partial insight: that an edge taxon should not be tagged as an attribute_type_id but rather, as a qualifier?

      species context qualifier:
        range: organism taxon
RichardBruskiewich commented 1 year ago

I suspect too that other "category" use cases will be specific qualifier "reified predicate" use case 2 above may be moot (don't have any examples) so I'll close this issue for now.