compound eye retinal cell is at an odd level of specifiity

[cmungall]

cell and ('part of' some ('photoreceptor array' and ('part of' some 'compound eye')))

in dmel, this is a grouping of eight photoreceptors, four cone cells, and two primary pigment cells

it was created to support

id: GO:1990010 name: compound eye retinal cell apoptotic process namespace: biological_process def: "Any apoptotic process in a compound eye retinal cell." [GOC:mtg_apoptosis, PMID:12021768] is_a: GO:1990009 ! retinal cell apoptotic process intersection_of: GO:0006915 ! apoptotic process intersection_of: occurs_in CL:0009001 ! compound eye retinal cell created_by: pr creation_date: 2012-12-12T10:21:07Z

note that FBbt doesn't have a class for this level of granularity AFAICT

looking at that pub and at https://www.nature.com/articles/4400767 I believe the cells that undergo PCD are the interommatidial retinal pigment precursors

in fbbt:

id: FBbt:00007369 name: interommatidial precursor cell def: "Precursor of the interommatidial cells (secondary and tertiary pigment cells). From approximately 27 hours after puparium formation (APF), 25% of these cells undergo programmed cell death and the rest differentiate. This produces a structured hexagonal lattice of six secondary pigment cells and three tertiary pigment cells surrounding each ommatidium. This lattice also incorporates three interommatidial bristles and is mature by 48 hours APF." []

I think the general principle here should be that IF CL is to add a term that is likely specific to an organismal lineage (e.g. here the paper is about Dmel), then use the reference ontology for that species - particularly for fly and worm, which is outside the normal area of expertise for CL.

it's not always clear that pseudogeneralization from fly or worm is a good idea vs just using those ontologies themselves, but for practical reasons it is good to extend the shoreline of CL to correspond to what other ontologies like GO need (as composite uberon+ssAOs have issues). But if it is done it should be anchored to an existing term like interommatidial

I think the simplest thing here is to rename http://purl.obolibrary.org/obo/CL_0009001 to interommatidial precursor cell (+ its defs) and we can then change the GO term label. Technically this is a definition change, but I think it's safe

or if we want to be more conservative, add a subclass and then we can change the GO term to use this

[github-actions[bot]]

This issue has not seen any activity in the past 6 months; it will be closed automatically in one year from now if no action is taken.