SAE needs to recognize WBbt (worm anatomy) terms for contextual information

[vanaukenk]

Hi,

We are not able to autocomplete on C. elegans anatomy terms in the SAE, for example to add cell type or anatomy to a CC annotation.

These terms are usually available for WB curators in the graphical interface.

Thx.

[vanaukenk]

@tmushayahama As with #26 - will this issue be addressed upon merging your code with Noctua-dev? Thx.

[tmushayahama]

@cmungall can you comment on what closure to include the worm anatomy terms

[cmungall]

This pull request will fix the issue: https://github.com/geneontology/go-ontology/pull/15327

previously the ontology did not know what wbbt terms were, they were floating in their own graph, unconnected. This bridges wbbt classes with uberon, and uberon is connected with the right upper level terms

however, we will need to validate the combination, cc @balhoff

[balhoff]

5 unsatisfiable classes:

• 'Z4.appaaaa nucleus'
• 'Z4.appaaap nucleus'
• 'dorsal nerve cord'
• 'head muscle'
• 'ventral nerve cord'

[balhoff]

I'm guessing peripheral nervous system and central nervous system don't make sense for nematodes?

[raymond91125]

@balhoff What does unsatisfiable mean? I would like to see how these 5 may be different from other anatomy terms and if I can fix them if there's anything incorrect/inconsistent.

[cmungall]

Moved here: https://github.com/obophenotype/uberon/issues/1405

@balhoff I only get unsat for DNC and VNC

@raymond91125 - a class if unsatisfiable if it is equivalent to the empty set, for example being a subclass of two disjoint classes. You may want to go through the tutorial here http://go-protege-tutorial.readthedocs.io/en/latest/Disjointness.html

See also http://uberon.github.io/downloads.html#bridge https://github.com/obophenotype/uberon/wiki/inter-anatomy-ontology-bridge-ontologies

[raymond91125]

Is there anything more I need/could do from the Worm Anatomy side to help getting WBbt in?

[vanaukenk]

@raymond91125 It looks like the old issues are fixed, but two new issues surfaced: 'buccal cavity' and 'intestinal lumen' are unsatisfiable classes. One solution to fix this is to make the 'C. elegans Cell and Anatomy' equivalent to 'anatomical entity' and ('part of' some Caenorhabditis) rather than 'anatomical structure' and ('part of' some Caenorhabditis)

[vanaukenk]

Note: it would be good to have a regular QC check that the Uberon - WBbt bridge gives no errors.

[balhoff]

The term mapping should be fixed with the next Uberon release.

[cmungall]

Note: it would be good to have a regular QC check that the Uberon - WBbt bridge gives no errors

Historically we've had Jenkins jobs for individual bridges, e.g.

https://build.berkeleybop.org/job/check-uberon-zfa/

but we never had resources to follow up when we needed changes upstream. We can revisit this.

For now I will put a check in the uberon release process. Perhaps I can coordinate with @raymond91125 on the reciprocal

[vanaukenk]

@cmungall The WBbt terms are working fine in Noctua now. I can close this ticket, but wanted to confirm with you that the part above about checking that the Uberon - WBbt bridge is not generating errors is in place. If not, should we move that to a separate ticket in Uberon and/or WB? Thx.