search

OBOFoundry / OBOFoundry.github.io

Metadata and website for the Open Bio Ontologies Foundry Ontology Registry
http://obofoundry.org
Other
160 stars 201 forks source link

Campaign to standardize synonym types #2450

Open cthoyt opened 9 months ago

cthoyt commented 9 months ago

This is an "epic" issue, just as a placeholder to reference various efforts towards standardizing synonym type definitions across the OBO Foundry

Background

These are the two precursor discussions in the OMO tracker where we talked about starting to encode synonym type definitions in OMO:

  1. https://github.com/information-artifact-ontology/ontology-metadata/issues/70
  2. https://github.com/information-artifact-ontology/ontology-metadata/issues/122

Updating OMO

Here are issues/PRs where we added standards to OMO:

Specifically, see the template: https://github.com/information-artifact-ontology/ontology-metadata/blob/master/src/templates/annotation_properties.tsv

Ontology Issues/Updates

Ontology updates/issues, where we updated an ontology to conform to the standard

Uberon

CL

ZFA

NCBITaxon

ENVO

XAO

PO

HP

ChEBI

cmungall commented 9 months ago

Thank you!!!

On Sat, Sep 23, 2023 at 4:00 AM Charles Tapley Hoyt < @.***> wrote:

This is an "epic" issue, just as a placeholder to reference various efforts towards standardizing synonym type definitions across the OBO Foundry

These are the two precursor discussions in the OMO tracker where we talked about starting to encode synonym type definitions in OMO:

  1. information-artifact-ontology/ontology-metadata#70 https://github.com/information-artifact-ontology/ontology-metadata/issues/70
  2. information-artifact-ontology/ontology-metadata#122 https://github.com/information-artifact-ontology/ontology-metadata/issues/122

Here are issues/PRs where we added standards to OMO:

Specifically, see the template: https://github.com/information-artifact-ontology/ontology-metadata/blob/master/src/templates/annotation_properties.tsv

Ontology updates/issues, where we updated an ontology to conform to the standard:

— Reply to this email directly, view it on GitHub https://github.com/OBOFoundry/OBOFoundry.github.io/issues/2450, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAAMMOLU3KKR35W45D4JS2LX326MBANCNFSM6AAAAAA5EDX7SQ . You are receiving this because you are subscribed to this thread.Message ID: @.***>

matentzn commented 7 months ago

I need some methodological help here, how I can weigh the arguments of @hoganwr on #148 ("universal or near universal substitutabilty") against yours "useful for X, eg Chebi"?

In other words, how will we decide, moving forward, what is a valid synonym type to avoid overloading the concept? In the current sense, even "first name" could become a synonym type. Can we have a clear definition of synonym type before we move on to add all these?

cthoyt commented 5 months ago

We are only two ontologies away from being done with my initial list! HPO needs some more hand-holding, and ChEBI will be possible if we can decide to push through the synonym types they need to encode different kinds of chemical synonyms.

There's still more to do. For example, PRO and NCBITaxon both require some more work since they have a big variety of synonyms, some of which could probably be ontologized differently (or really would never make sense as a "standardized" or "externalized" synonym). PO is misusing synonyms to denote languages. There are probably other things floating around too!

cthoyt commented 4 months ago

I need some methodological help here, how I can weigh the arguments of @hoganwr on #148 ("universal or near universal substitutabilty") against yours "useful for X, eg Chebi"?

I want to counter this wrt the BRAND_NAME discussion, since many ontologies import chebi and therefore accidentally mint their own version, I think that there is always a case to be made for externalizing synonym types, and anyone who wants to be a stakeholder can participate in the discussion board here when someone proposes one

hoganwr commented 4 months ago

It occurs to me here that there are term "types", and there are "synonym" types. What are we discussing here specifically?

Brand name is suitable as a term type, but it is not a good synonym type.

Bill

On Wed, Feb 21, 2024 at 5:54 AM Charles Tapley Hoyt < @.***> wrote:

I need some methodological help here, how I can weigh the arguments of @hoganwr https://github.com/hoganwr on #148 https://github.com/OBOFoundry/OBOFoundry.github.io/issues/148 ("universal or near universal substitutabilty") against yours "useful for X, eg Chebi"?

I want to counter this wrt the BRAND_NAME discussion, since many ontologies import chebi and therefore accidentally mint their own version, I think that there is always a case to be made for externalizing synonym types, and anyone who wants to be a stakeholder can participate in the discussion board here when someone proposes one

— Reply to this email directly, view it on GitHub https://github.com/OBOFoundry/OBOFoundry.github.io/issues/2450#issuecomment-1956489775, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAJR55QEZOJSNVZAF5U4SRLYUXN7ZAVCNFSM6AAAAAA5EDX7SSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNJWGQ4DSNZXGU . You are receiving this because you were mentioned.Message ID: @.***>