adding GCI to infer part-of edges for morphogenesis. Addresses #12658

[cmungall]

No description provided.

[cmungall]

@balhoff do we have a way of testing the generative capacities?

for now I am just running by hand:

dosdp-tools generate --obo-prefixes true --ontology go-edit.obo --catalog catalog-v001.xml --template ../design_patterns/x_morphogenesis.yaml --infile ../design_patterns/x_morphogenesis.tsv --outfile foo.owl

This is producing two flavors of GCI as expected - next step is to hook up materialize

[balhoff]

I think we'll want to create separate YAMLs just to generate the GCIs, right? Otherwise we're going to create extra labels, etc. I'm doing something like this for Biochebi: https://github.com/geneontology/go-ontology/blob/master/src/biochebi/biochebi_gcis.yaml

[balhoff]

We are already materializing 'part of', so if we import the GCIs it should just work.

[balhoff]

I did a check and find that including the GCIs in the DP causes the DP match query to find no results (since the GCI doesn't exist in the ontology). I think we should keep the GCIs in a separate generative file, since we use these files for matching.

[balhoff]

@cmungall I made several additions to the PR, to separate out the GCI pattern, and add generation steps to the makefile. What is a test case part_of we should be looking for?

[balhoff]

Two resulting term equivalences:

• regulation of cell morphogenesis == regulation of cell morphogenesis involved in differentiation
• cell morphogenesis == cell morphogenesis involved in differentiation

[balhoff]

regulation of cell morphogenesis involved in differentiation EquivalentTo regulation of cell morphogenesis

• regulation of cell morphogenesis involved in differentiation EquivalentTo biological regulation and (regulates some cell morphogenesis involved in differentiation)
  • cell morphogenesis involved in differentiation EquivalentTo cell morphogenesis and (part of some cell development)
    • cell morphogenesis EquivalentTo anatomical structure morphogenesis and (results in morphogenesis of some cell)
    • cell development EquivalentTo biological_process and (results in development of some cell)
• anatomical structure morphogenesis and (results in morphogenesis of some cell) SubClassOf part of some
  (anatomical structure development and (results in development of some cell))
• anatomical structure development SubClassOf developmental process
• developmental process SubClassOf biological_process
• regulation of cell morphogenesis EquivalentTo biological regulation and (regulates some cell morphogenesis)

[cmungall]

OK, let's leave the GCIs in a seperate file for now, but these are already quite repetitious, involving repetition of information from other patterns. It might be an idea to have an option on match to selectively match on equivalence axioms, since that is the main use case

[cmungall]

Shall we make a separate issue with the proposal to merge cell morphogenesis involved in differentiation? We can include the explanation there (we don't need to discuss the regulation since it trivially follows)

[balhoff]

Shall we make a separate issue with the proposal to merge cell morphogenesis involved in differentiation? We can include the explanation there (we don't need to discuss the regulation since it trivially follows)

#24299