OBOFoundry / COB

An experimental ontology containing key terms from Open Biological and Biomedical Ontologies (OBO)
https://obofoundry.github.io/COB
Creative Commons Zero v1.0 Universal
39 stars 8 forks source link

Generalised [spatial, causal, ...] relations #36

Open pbuttigieg opened 5 years ago

pbuttigieg commented 5 years ago

Will Core (CORE?) include relations that can be generalised beyond biology? ENVO would need geospatial relations (BSPO-like) and causal algebra that isn't bound to a biological entity. Many aren't, but it pops up some times.

Could we ensure that the top level stuff is very general?

pbuttigieg commented 5 years ago

Not defined yet, but a good example of something we'd use in ENVO's geological processes: http://purl.obolibrary.org/obo/RO_0002297 (results in formation of)

cmungall commented 5 years ago

This is a good question. ENVO has been piggybacking off of GO's relations many of which are purportedly general but there are good and bad reasons for having parallel sets for different domains.

On Tue, Jul 30, 2019 at 10:03 AM Pier Luigi Buttigieg < notifications@github.com> wrote:

Not defined yet, but a good example of something we'd use in ENVO's geological processes: http://purl.obolibrary.org/obo/RO_0002297 (results in formation of)

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/OBOFoundry/Experimental-OBO-Core/issues/36?email_source=notifications&email_token=AAAMMOKTTXHMFAXAZ23Z3LLQCBXYNA5CNFSM4IH6M7PKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3EUHNI#issuecomment-516506549, or mute the thread https://github.com/notifications/unsubscribe-auth/AAAMMOKQ6Y37NJS7D4YX7JTQCBXYNANCNFSM4IH6M7PA .

bpeters42 commented 5 years ago

Good point Pier. I do think every relation in Core that does not explicitly have a domain / range that restricts it to something biological should be applicable to non-biological domains. The "Results in formation of" relation you point to is a good example that should not make it into the core unless it is much better defined (and maybe gets a new label). But we should have the parent relation "has output" in the Core, which is applicable to geological (and any other) processes as is.

On Sun, Aug 4, 2019 at 11:07 AM Chris Mungall notifications@github.com wrote:

This is a good question. ENVO has been piggybacking off of GO's relations many of which are purportedly general but there are good and bad reasons for having parallel sets for different domains.

On Tue, Jul 30, 2019 at 10:03 AM Pier Luigi Buttigieg < notifications@github.com> wrote:

Not defined yet, but a good example of something we'd use in ENVO's geological processes: http://purl.obolibrary.org/obo/RO_0002297 (results in formation of)

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub < https://github.com/OBOFoundry/Experimental-OBO-Core/issues/36?email_source=notifications&email_token=AAAMMOKTTXHMFAXAZ23Z3LLQCBXYNA5CNFSM4IH6M7PKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3EUHNI#issuecomment-516506549 , or mute the thread < https://github.com/notifications/unsubscribe-auth/AAAMMOKQ6Y37NJS7D4YX7JTQCBXYNANCNFSM4IH6M7PA

.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/OBOFoundry/Experimental-OBO-Core/issues/36?email_source=notifications&email_token=ADJX2IXRVUK6QMEC56ANAPLQC4LHHA5CNFSM4IH6M7PKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3QGWNQ#issuecomment-518023990, or mute the thread https://github.com/notifications/unsubscribe-auth/ADJX2IQYBBTB2TU3MKKQCADQC4LHHANCNFSM4IH6M7PA .

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