Unable to query based on class type, e.g., rdf:Class or owl:Class

[tekrajchhetri]

Currently, we cannot perform queries to retrieve all classes defined as rdfs:Class or owl:Class. For instance, while GeneAnnotation is a class, the current annotations do not explicitly indicate this, making it impossible to identify GeneAnnotation as a class in such queries.

Below is a snippet of the current version of annotated data (in TTL representation). As we can clearly see that its lacking the explicit indication of the class.

@prefix NCBIAssembly: <https://www.ncbi.nlm.nih.gov/assembly/> .
@prefix NCBIGene: <http://identifiers.org/ncbigene/> .
@prefix bican: <https://identifiers.org/brain-bican/vocab/> .
@prefix biolink: <https://w3id.org/biolink/vocab/> .
@prefix dcterms: <http://purl.org/dc/terms/> .
@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .
@prefix schema1: <http://schema.org/> .
@prefix owl: <http://www.w3.org/2002/07/owl#> .


bican:000015fd3d6a449b47e75651210a6cc74fca918255232c8af9e46d077034c84d a bican:GeneAnnotation ;
    rdfs:label "LOC106504536" ;
    schema1:identifier "106504536" ;
    bican:molecular_type "protein_coding" ;
    bican:referenced_in bican:d5c45501b3b8e5d8b5b5ba0f4d72750d8548515c1b00c23473a03a213f15360a ;
    biolink:category bican:GeneAnnotation ;
    biolink:in_taxon bican:7d54dfcbd21418ea26d9bfd51015414b6ad1d3760d09672afc2e1e4e6c7da1dd ;
    biolink:in_taxon_label "Sus scrofa" ;
    biolink:symbol "LOC106504536" ;
    biolink:xref NCBIGene:106504536 .

To address this issue, we need the annotation in the following format.

@prefix NCBIAssembly: <https://www.ncbi.nlm.nih.gov/assembly/> .
@prefix NCBIGene: <http://identifiers.org/ncbigene/> .
@prefix bican: <https://identifiers.org/brain-bican/vocab/> .
@prefix biolink: <https://w3id.org/biolink/vocab/> .
@prefix dcterms: <http://purl.org/dc/terms/> .
@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .
@prefix schema1: <http://schema.org/> .
@prefix owl: <http://www.w3.org/2002/07/owl#> .

bican:GeneAnnotation a owl:Class ;
    rdfs:label "Gene Annotation" ;
    rdfs:comment "A class representing gene annotations." .

bican:000015fd3d6a449b47e75651210a6cc74fca918255232c8af9e46d077034c84d a bican:GeneAnnotation ;
    rdfs:label "LOC106504536" ;
    schema1:identifier "106504536" ;
    bican:molecular_type "protein_coding" ;
    bican:referenced_in bican:d5c45501b3b8e5d8b5b5ba0f4d72750d8548515c1b00c23473a03a213f15360a ;
    biolink:category bican:GeneAnnotation ;
    biolink:in_taxon bican:7d54dfcbd21418ea26d9bfd51015414b6ad1d3760d09672afc2e1e4e6c7da1dd ;
    biolink:in_taxon_label "Sus scrofa" ;
    biolink:symbol "LOC106504536" ;
    biolink:xref NCBIGene:106504536 .

In this version, we now have explicit semantics encoded, enabling us to query (see below) by rdfs:Class.

bican:GeneAnnotation a owl:Class ;
    rdfs:label "Gene Annotation" ;
    rdfs:comment "A class representing gene annotations." .

Query:

PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#> 

SELECT DISTINCT ?class 
WHERE {
    ?class a rdfs:Class
}

[satra]

perhaps test out the gen-jsonld for the linkml model itself. i.e. what this needs is a conversion of the model itself, not data encoded in the model. perhaps there is something that exports the model as owl as well.

[djarecka]

I could generate an owl representation of the model and see how it looks like. I will be testing newer version of linkml, so can add it as well.

[satra]

also relatedly, this is where we should pull in any related/relevant ontologies.

[tekrajchhetri]

Yes, we can pull the relevant ontologies. I was not sure if it would be the best approach as in some cases we might be using just a class.