Replace CARO in domain/range with UBERON terms

[anitacaron]

Related to obophenotype/cell-ontology#1867

A new subset was defined in UBERON to replace CARO ontology.

@balhoff, could you please give instructions on how to use the subset in this case?

[balhoff]

I've just opened a relevant issue for Uberon: obophenotype/uberon#2835

But also I think just referencing the terms and doing an ordinary import module extraction via ODK should work.

[ddooley]

These are the needed CARO - UBERON substitutions, and we'll need an UBERON import module for them. Jim will provide mapped terms, Damion will do the replacement; Anita, can you prepare the UBERON import file?

[balhoff]

Most of these are easy, except for two:

• anatomical entity
• connected anatomical structure -> anatomical structure
• immaterial anatomical entity
• material anatomical entity
• multicellular anatomical structure
• neuron projection bundle
• organism or virus or viroid --> we will use COB:organism

[balhoff]

@gouttegd I think you have grappled with 'connected anatomical structure' recently. We don't seem to have a direct replacement from Uberon. Do you think we can just use 'material anatomical entity' in its place in RO?

[balhoff]

@cmungall any thoughts on replacing organism or virus or viroid?

[bpeters42]

In COB, this CARO term is already mapped to 'organism' with 'cellular organism' as a child. This has been a very long discussion in the past, and we should consider it solved.

…

On Tue, May 23, 2023 at 10:12 AM Jim Balhoff ***@***.***> wrote: @cmungall <https://github.com/cmungall> any thoughts on replacing organism or virus or viroid <http://purl.obolibrary.org/obo/CARO_0001010>? — Reply to this email directly, view it on GitHub <#695 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADJX2IRDCYUG3URF5233K5DXHTVX7ANCNFSM6AAAAAAVZHAUOQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

-- Bjoern Peters Professor La Jolla Institute for Immunology 9420 Athena Circle La Jolla, CA 92037, USA Tel: 858/752-6914 Fax: 858/752-6987 http://www.liai.org/pages/faculty-peters

[balhoff]

Thanks for pointing that out @bpeters42 — I guess that means that #716 is a prerequisite, so we can use that COB term.

[gouttegd]

@balhoff

I think you have grappled with 'connected anatomical structure' recently

I’ve had this misfortune, yes. :D

We don't seem to have a direct replacement from Uberon. Do you think we can just use 'material anatomical entity' in its place in RO?

I think it can actually be replaced by Uberon’s anatomical structure. Though it’s not obvious from the label,¹ the term has a text definition that clearly states that it refers to a connected structure (“Material anatomical entity that is a single connected structure […]”), it has connected anatomical structure as an exact synonym, and it is cross-referenced to CARO’s connected anatomical structure.

Footnotes

1. If I remember correctly, the fact that the label does not make it obvious the term refers to something connected was the cause of most of the problems I had to deal with, because people had used the term for things that were not connected, probably because they didn’t check the textual definition. ↩

[balhoff]

@gouttegd thanks! I updated the list above.

[ddooley]

@anitacaron is it possible to get the UBERON import module pull request first, then I can do the substitution.

[ddooley]

@bpeters42 I open up COB and see "organism" is CARO:0001010 . Is that intentional? It would mean CARO is still needed in RO! It seems cob-to-external indicates equivalency to both OBI and CARO organism.

[bpeters42]

See the mapping file in COB; organism maps both to OBI, and one to CARO. The CARO class was created to align with the one in OBI. If CARO gets obsoleted, we can remove the CARO mapping, but still have the OBI one (which has massive external use, even though OBI shouldn't be a source for organism terms). COB:0000022 organism owl:equivalentClass OBI:0100026 organism COB:0000022 organism owl:equivalentClass CARO:0001010 organism or virus or viroid

…

On Tue, May 23, 2023 at 1:42 PM Damion Dooley ***@***.***> wrote: @bpeters42 <https://github.com/bpeters42> I open up COB and see "organism" is CARO:0001010 . Is that intentional? It would mean CARO is still needed in RO! — Reply to this email directly, view it on GitHub <#695 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADJX2ISFA4VGEM66I4TRKC3XHUOLHANCNFSM6AAAAAAVZHAUOQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>

-- Bjoern Peters Professor La Jolla Institute for Immunology 9420 Athena Circle La Jolla, CA 92037, USA Tel: 858/752-6914 Fax: 858/752-6987 http://www.liai.org/pages/faculty-peters

[ddooley]

Ok, I'll have replacement list switch organism to OBI:0100026 .

[balhoff]

@ddooley Uberon import is available now.

[ddooley]

One question about that. Whats the difference in ro-odk.yaml in products: section between module type minimal vs filter? Why is uberon a filter module_type whereas pato is a minimal one? I tried switching the uberon one and couldn't really see a difference. I made the new obi module a minimal one, and have imported it.

- id: uberon
  module_type: filter
- id: obi
  make_base: TRUE
  module_type: minimal

I realize COB might be a source of terms, but then we get into task of needing to have COB load up all the ones referenced for import in RO.

[matentzn]

@ddooley there used to be a bigger difference between filter and minimal, right now the only one is that minimal will not automatically delete object properties if they are not included in the seed, while filter strictly only includes what is in the seed. For most cases, we should use filter.

[wdduncan]

Closed via #722.