dosumis
commented
1 year ago Can the range restriction of has soma location be relaxed to a material anatomical entity with no assertions regarding its “connectedness”?
Yes please. I think this is also incompatible with the property chain over part_of as connected structures can be part of disconnected ones.
cmungall
has soma location (RO:0002100) has a range restriction to connected anatomical structure (CARO:0000003).
Combined with PATO’s general class axiom enforcing a disjunction between
'has characteristic' some 'maximally connected'and'has characteristic' some disconnected, this prevents any neuron from having its soma located in any kind of anatomical structure that is not maximally connected.When merging FBbt with Uberon/CL, this causes ~600 unsatisfiable classes due to a large number of fly neurons that are said in FBbt to have their soma located in some kind of sensory clusters, because anatomical clusters are disconnected entities.
Can the range restriction of has soma location be relaxed to a material anatomical entity with no assertions regarding its “connectedness”?
If not, what would be the appropriate relation to state that a neuron has its soma within a disconnected structure such as a cluster? part of is not ideal since the neuron may not entirely be located within the cluster – its neurites may extend beyond.