obophenotype / cell-ontology

An ontology of cell types
https://obophenotype.github.io/cell-ontology/
Creative Commons Attribution 4.0 International
142 stars 49 forks source link

[Class hierarchy] #2621

Closed tnetx closed 2 hours ago

tnetx commented 17 hours ago

CL term Seen in https://github.com/obophenotype/cell-ontology/blob/v2024-09-26/cl.obo

id: CL:0000163
name: endocrine cell
def: "A cell of an endocrine gland, ductless glands that secrete substances which are released directly into the circulation and which influence metabolism and other body functions." [MESH:D055098]
subset: general_cell_types_upper_slim
subset: human_reference_atlas
synonym: "endocrinocyte" EXACT []
xref: FMA:83809
is_a: CL:0000151 ! secretory cell
is_a: CL:0000164 {gci_relation="part_of", gci_filler="UBERON:0000945"} ! enteroendocrine cell
is_a: CL:0000164 {gci_filler="UBERON:0000160", gci_relation="part_of"} ! enteroendocrine cell
is_a: CL:0000164 {gci_filler="UBERON:0001264", gci_relation="part_of"} ! enteroendocrine cell
relationship: part_of UBERON:0000949 ! endocrine system
relationship: present_in_taxon NCBITaxon:9606 ! Homo sapiens

[Term]
id: CL:0000164
name: enteroendocrine cell
def: "An endocrine cell that is located in the epithelium of the gastrointestinal tract or in the pancreas." [GOC:tfm, SANBI:mhl]
xref: BTO:0003865
xref: FMA:62930
xref: MESH:D019858
is_a: CL:0000150 ! glandular epithelial cell
is_a: CL:0000163 ! endocrine cell

Suggested revision of class hierarchy

There seems to be a circular parent-child hierarchy between CL:0000163 (endocrine cell) and CL:0000164 (enteroendocrine cell). Are the is_a relationships from CL:0000163 to CL:0000164 autogenerated? It seems that only CL:0000164 should be "is_a" CL:0000163.

dosumis commented 2 hours ago

These are OWL GCIs & are doing useful work.

id: CL:0000163 name: endocrine cell is_a: CL:0000164 {gci_relation="part_of", gci_filler="UBERON:0000945"} ! enteroendocrine cell

=>

OWL: endocrine cell and part_of some stomach subClassof enteroendocrine cell English: All endocrine cells that are part of the stomach are enteroendoctine cells

The representation in OBO is potentially misleading.

tnetx commented 2 hours ago

@dosumis I do indeed think it's misleading. Would it be preferable to state it as relationship: part_of: UBERON:0000945 on the enteroendocrine Term instead?

id: CL:0000164
name: enteroendocrine cell
is_a: CL:0000163 ! endocrine cell
...
relationship: part_of: UBERON:0000945 ! stomach
dosumis commented 2 hours ago

It is not true that all enteroendocrine cells are part of the stomach, some are part of the pancreas, so "relationship: part_of: UBERON:0000945 ! stomach" would be incorrect .The OWL expression allows us to assert both part relationships and use the3m to autoclassify endocrine cells as enteroendocrine.

The issue is not with the assertions but with confusion caused by the OBO representation.

Could you let us know what you are trying to achieve? You may not need to use the OBO file.