Changing mapping that is causing unsat on Uberon

[anitacaron]

• DMBA_16271 is EquivalentTo lateral tuberal nucleus that is (part of some Mus musculus),
• but lateral tuberal nucleus is in taxon some Primates.

This is a simple mapping error. In Uberon lateral tuberal nucleus is explicitly intended to refer to a structure in human and higher primates, there is a distinct term intended to refer to a similar structure in rodents: tuberal nucleus (sensu Rodentia), where a comment explains:

While lateral tuberal nucleus is often used interchangeably with tuberal nucleus in rodents, lateral tuberal nucleus is specific to humans and higher primates. In some cases, lateral tuberal nucleus is thought to be part of the tuberal nucleus in rodents, and could possibly be homologous to human tuberal region instead.

Of note, this erroneous mapping comes from one of the custom bridges we have with the Allen Mouse Brain Atlas, which are maintained elsewhere and simply used as they come in Uberon. So the fix needs to happen in the remote source.

Originally posted by @gouttegd in obophenotype/uberon#2928 (comment)

[anitacaron]

cc @dosumis

[dosumis]

We are moving the location of these mappings to the BICAN repos - splitting to separate repos so we can provide a PURL for each. Here's DMBA: https://github.com/brain-bican/developing_mouse_brain_atlas_ontology/tree/main @hkir-dev is responsible for the move.

@gouttegd - what is the build process that is finding this? Looks like we need to review and move to the new repos. DO we need to fix in more than one place in this case given the current assumptions about builds?

[gouttegd]

The bridge files in this repo are referenced from the Uberon Makefile, lines 1254–1272.

[dosumis]

@hkir-dev - can we switch to equivalent files on the new DMBA repo? We can then update mapping there.

[hkir-dev]

Equivalent files are:

• https://raw.githubusercontent.com/brain-bican/mouse_brain_atlas_ontology/main/src/ontology/new-bridges/new-uberon-bridge-to-mba.owl
• https://raw.githubusercontent.com/brain-bican/developing_mouse_brain_atlas_ontology/main/src/ontology/new-bridges/new-uberon-bridge-to-dmba.owl

I can create a pull request to change locations. But we changed namespace of the entities as well. Such as http://purl.obolibrary.org/obo/DMBA_112892315 to https://purl.brain-bican.org/ontology/dmbao/DMBA_112892315

Will this namespace change cause any problem?

[dosumis]

Will this namespace change cause any problem?

@gouttegd ?

[gouttegd]

I don’t foresee any issue with that at least within Uberon. DMBA entities are not referenced from anywhere else than the bridge itself, so the namespace change should not require to update anything.

[gouttegd]

I’ve updated the Uberon Makefile to pull the MBA/DMBA bridges from the new indicated location (in a private branch, for now). Can the issue about DMBA:16271 be fixed in https://github.com/brain-bican/developing_mouse_brain_atlas_ontology?

[dosumis]

brain-bican/developing_mouse_brain_atlas_ontology#1

[matentzn]

@dosumis can we please prioritise this issue here soon? Its a bit disruptive to have this in the Uberon pipeline like this.. :P Thanks!

[hkir-dev]

The bridge file updated at https://github.com/brain-bican/developing_mouse_brain_atlas_ontology/blob/main/src/ontology/new-bridges/new-uberon-bridge-to-dmba.owl
@gouttegd do you still have that branch or should I make a new PR to Uberon for that?

[gouttegd]

@hkir-dev Not sure what you mean by “do you still have that branch”.

Uberon is configured to fetch the bridge from that branch, if that’s what you mean. So if the issue has been fixed there, we will automatically get the fix the next time we update our external resources. :)