"parents" relation in the dump data reflects the hierarchy of the ontology. So we have CL classes, multiple parents etc.
But to build sunburst diagram, we need the hierarchy mentioned in the source data. Should we provide another hierarchy that reflects the simplified "subcluster_of" hierarchy.
From Jim:
As a spot test, I manually traced through L6 IT ancestors
I’ve attached a report of the 42 instances where we do not find a parent whose rank is class, subclass && the taxonomy does not match.
hierarchy-lineage-report.txt
"parents" relation in the dump data reflects the hierarchy of the ontology. So we have CL classes, multiple parents etc.
But to build sunburst diagram, we need the hierarchy mentioned in the source data. Should we provide another hierarchy that reflects the simplified "subcluster_of" hierarchy.
From Jim: As a spot test, I manually traced through L6 IT ancestors
test case: L6 IT | http://purl.obolibrary.org/obo/PCL_0011192 expected: Glutamatergic | http://purl.obolibrary.org/obo/PCL_0011579
id: http://purl.obolibrary.org/obo/PCL_0011192 prefLabel: [ ‘L6 IT' ] parents: [ 'http://purl.obolibrary.org/obo/PCL_0011182' ] ↓ id: http://purl.obolibrary.org/obo/PCL_0011182 prefLabel: [ ‘IT projecting’ ] parents: [ ‘http://purl.obolibrary.org/obo/CL_4023008’ ] ↓ id: http://purl.obolibrary.org/obo/CL_4023008 label: [ ‘intratelencephalic-projecting glutamatergic cortical neuron’ ] parents: [] NO PARENT
I’ve attached a report of the 42 instances where we do not find a parent whose rank is class, subclass && the taxonomy does not match. hierarchy-lineage-report.txt