Release 2022 03 31 (easier review)

[matentzn]

No description provided.

[shawntanzk]

removed anatomical entity doesn seem a big deal, as they have is_a axioms to something that is basically an anatomical entity, im guessing that this is because of the changes in bridge equivalence to caro - don't think its a huge deal

[shawntanzk]

Copying this over:

Things I would like to get signed off/source found before committing to the release:

1. The whole lot of epithelium related stuff that are now just anatomical entity, heaps of changes with this, would like to know the source of the changes
2. The massive removal of neural crest related axioms
3. The huge removal of subclassof anatomical entity (they do seem to be anatomical entities so not sure why this is happening)

[shawntanzk]

for the epithelium stuff, it might be this: 5a65039 in which case it should be fine, but its weird that it didnt show up in the logical diff?

[shawntanzk]

might be of concern, left is this branch, right is last release - GCIs seem to be removed

[shawntanzk]

I'm looking at a few classes with 'structure with developmental contribution from neural crest' - and it seems that they can be inferred some other way, so it makes sense that those axioms are removed. The question I have is why is it happening this time?

[matentzn]

This is super weird! So many GCIs removed does not smell good. Can we somehow figure out which file they were supposed to be in? originally? (The GCIs cannot be in Uberon..). Is it possible they came back through some import in the past?

[shawntanzk]

This is super weird! So many GCIs removed does not smell good. Can we somehow figure out which file they were supposed to be in? originally? (The GCIs cannot be in Uberon..). Is it possible they came back through some import in the past?

oh that explains it, sorry lol, was using the edit file, should probably have used the built uberon >.< just checked all good

[shawntanzk]

left to do: see if epithelium stuff is also just removing things that can be inferred

[shawntanzk]

It seems like this new release removed many axioms that can technically be inferred:

I'm not sure why this is done as the inferences are pretty complex (like you need to know what you are looking for for that) Need input on how to proceed.

[shawntanzk]

Potential reason for massive changes: obophenotype/cell-ontology#1532

[balhoff]

These property chains were removed on purpose: #1808

[shawntanzk]

strange that the changes are only showing up in this release? I'm guessing because of all the import dependencies, it took awhile to finally be clean?

[balhoff]

I agree, it sure took a while! But the imports have been a tangle. There are some axioms that only live in imports and get passed back and forth.

[matentzn]

@shawntanzk can you do the final review here and approve!

[matentzn]

I approve this!

[matentzn]

For future reference, this is yet undadressed: #2382