search

obophenotype / brain_data_standards_ontologies

A repository for co-ordinating work on ontologies for the Brain Data Standards Project
Apache License 2.0
11 stars 3 forks source link

Switching to pairwise matching for cross-species #194

Closed shawntanzk closed 2 years ago

shawntanzk commented 3 years ago

Schema: see https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7672696/ We should use homologous_to (“in historical homology relationship with” (RO:HOM0000007)

Cell_set_aligned_alias from mouse/human/marmoset should be RO:HOM0000007 the same name as Cell_set_aligned_alias of each other

This will eliminate the need for X-speices taxonomy which we will "switch off" for now - however if biologist like Ed would like it in, we should be able to easily switch it on again

shawntanzk commented 3 years ago

These issues are left over from checks on switching mouse to nomenclature file, however they will no longer be an issue with Euarchontoglires terms being removed - will need to check that pairwise matches work though:

VLMC in mouse not under VLMC MOp (Euarchontoglires) Micro MOp (mouse) not under Microglia/PVM MOp

shawntanzk commented 3 years ago

Jeremy will be merging aligned alias to one column and everything should match up that way for pairwise. Will wait for that to happen before this implementation

shawntanzk commented 3 years ago

Currently pairwise matching uses RO:HOM0000007 which is an object property to represent homologous types As it is an object property and not annotation, everything under it gets that property too This seems wrong to me as the child nodes would not be homologous to the above, however, this is unavoidable as an object property. We cannot use this as an annotation property as an object property cannot be an annotation property (https://github.com/oborel/obo-relations/issues/101) There does not appear to be any annotation property suitable at the moment too

shawntanzk commented 2 years ago

Implemented