search

obi-ontology / obi

The Ontology for Biomedical Investigations
http://obi-ontology.org
Creative Commons Attribution 4.0 International
75 stars 25 forks source link

Bad equivalent classes 'lower respiratory tract specimen' and 'lower respiratory tract aspirate specimen'. #1476

Closed jamesaoverton closed 1 year ago

jamesaoverton commented 2 years ago

I noticed this when looking into #1475. These two classes are being inferred as equivalent:

Both are defined as equivalent to "specimen and is_specified_output_of some (collecting specimen with swab and has_specified_input some lower respiratory tract)".

One option is to make 'lower respiratory tract aspirate specimen' output of an 'aspiration specimen collection' rather than 'collecting specimen with swab'. Another option is to include some sort of aspiration process term. There may be other good options that I'm not thinking of.

What would you prefer @linikujp ?

bpeters42 commented 2 years ago

I thought we had a template for specimens?

On Mon, Jan 10, 2022 at 9:26 AM James A. Overton @.***> wrote:

I noticed this when looking into #1475 https://github.com/obi-ontology/obi/issues/1475. These two classes are being inferred as equivalent:

Both are defined as equivalent to "specimen and is_specified_output_of some (collecting specimen with swab and has_specified_input some lower respiratory tract)".

One option is to make 'lower respiratory tract aspirate specimen' output of an 'aspiration specimen collection' http://purl.obolibrary.org/obo/OBI_0002785 rather than 'collecting specimen with swab'. Another option is to include some sort of aspiration process term. There may be other good options that I'm not thinking of.

What would you prefer @linikujp https://github.com/linikujp ?

— Reply to this email directly, view it on GitHub https://github.com/obi-ontology/obi/issues/1476, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADJX2IS7MOEDHQL6MEGCWPLUVMJCRANCNFSM5LUGYTDQ . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

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

jamesaoverton commented 2 years ago

These two are in biobank-specimens.tsv template which has columns for 'gross anatomical part' and 'swabbed surface'. These two currently use the 'swabbed surface' column.

bpeters42 commented 2 years ago

OK: it looks like we need to generalize 'sample being taken from surface of X', and or 'from material in cavity'. There are other ways than doing it with a swab. In the case of an aspirate, it is drawing a fluid contained in X, here in the airways. This is worth also discussing with the anatomy ontologies, so pinging @cmungall

ddooley commented 2 years ago

How about:

lower respiratory tract specimen subClassOf: 'is specified output of' some ('specimen collection process' and 'occurs in' some 'lower respiratory tract')

lower respiratory tract aspirate specimen subClassOf: 'is specified output of' some ('aspiration specimen collection' and 'occurs in' some 'lower respiratory tract')

linikujp commented 2 years ago

I like Damion's proposal!

On Mon, May 2, 2022 at 10:55 AM Damion Dooley @.***> wrote:

How about:

lower respiratory tract specimen subClassOf: 'is specified output of' some ('specimen collection process' and 'occurs in' some 'lower respiratory tract')

lower respiratory tract aspirate specimen subClassOf: 'is specified output of' some ('aspiration specimen collection' and 'occurs in' some 'lower respiratory tract')

— Reply to this email directly, view it on GitHub https://github.com/obi-ontology/obi/issues/1476#issuecomment-1114997128, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACCPEPMEZTSUM2HV5B7DVJTVH7UFDANCNFSM5LUGYTDQ . You are receiving this because you were mentioned.Message ID: @.***>

ddooley commented 2 years ago

OBI May 2 call: Will add new axiom column containing ['is specified output of' some ....] to biobank template to switch to above subClassOf items.

cmungall commented 2 years ago

so your solution is to weaken EC axioms to SC? Not wrong but you lose a lot of useful inference.

why not just use http://purl.obolibrary.org/obo/OBI_0002785 as your genus?

ddooley commented 2 years ago

Yes, downgrade to SC.I recall discussion seemed to arrive at that it is hard to define aspirate axiomatically.

"lower respiratory tract aspirate specimen" is a specimen so can't have genius aspiration specimen collection process?

bpeters42 commented 2 years ago

I missed the call today, so I may very well be of base. But I think we can stick to equivalent class axioms for these specimens once we have implemented "OBI May 2 call: Will add new axiom column containing ['is specified output of' some ....] to biobank template [...]

While it is hard to axiomatically define 'aspiration' to distinguish it from other specimen collections, we are not talking about making 'aspiration' a defined class. It seems to me that defining 'aspirate specimen' as specimen that are collected with a 'specimen collection by aspiration' process, and then having the combinations like 'fine needle aspirate specimen' / 'lower respiratory tract aspirate specimen' and the like should work.

On Mon, May 2, 2022 at 10:10 AM Damion Dooley @.***> wrote:

Yes, downgrade to SC.I recall discussion seemed to arrive at that it is hard to define aspirate axiomatically.

"lower respiratory tract aspirate specimen" is a specimen so can't have genius aspiration specimen collection http://purl.obolibrary.org/obo/OBI_0002785 process?

— Reply to this email directly, view it on GitHub https://github.com/obi-ontology/obi/issues/1476#issuecomment-1115133562, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADJX2ITSIDS2VL6TNCQNKSLVIAD7DANCNFSM5LUGYTDQ . You are receiving this because you commented.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

cmungall commented 2 years ago

Agree with Bjoern

And in general if you have dozens of classes

Foo a Foo b Foo c

Etc where a b and c are all classes

Then you can just make simple yet powerful logical definitions using foo as a fixed element, don’t need to worry about foo too much, only that it means the same thing in all cases

Weakening compositional classes to SC axioms is usually an anti pattern

On Mon, May 2, 2022 at 6:19 PM bpeters42 @.***> wrote:

I missed the call today, so I may very well be of base. But I think we can stick to equivalent class axioms for these specimens once we have implemented "OBI May 2 call: Will add new axiom column containing ['is specified output of' some ....] to biobank template [...]

While it is hard to axiomatically define 'aspiration' to distinguish it from other specimen collections, we are not talking about making 'aspiration' a defined class. It seems to me that defining 'aspirate specimen' as specimen that are collected with a 'specimen collection by aspiration' process, and then having the combinations like 'fine needle aspirate specimen' / 'lower respiratory tract aspirate specimen' and the like should work.

  • Bjoern

On Mon, May 2, 2022 at 10:10 AM Damion Dooley @.***> wrote:

Yes, downgrade to SC.I recall discussion seemed to arrive at that it is hard to define aspirate axiomatically.

"lower respiratory tract aspirate specimen" is a specimen so can't have genius aspiration specimen collection http://purl.obolibrary.org/obo/OBI_0002785 process?

— Reply to this email directly, view it on GitHub <https://github.com/obi-ontology/obi/issues/1476#issuecomment-1115133562 , or unsubscribe < https://github.com/notifications/unsubscribe-auth/ADJX2ITSIDS2VL6TNCQNKSLVIAD7DANCNFSM5LUGYTDQ

. You are receiving this because you commented.Message ID: @.***>

-- Bjoern Peters Professor La Jolla Institute for Immunology 9420 Athena Circle https://www.google.com/maps/search/9420+Athena+Circle+%0D%0D%0ALa+Jolla,+CA+92037,+USA?entry=gmail&source=g La Jolla, CA 92037, USA https://www.google.com/maps/search/9420+Athena+Circle+%0D%0D%0ALa+Jolla,+CA+92037,+USA?entry=gmail&source=g Tel: 858/752-6914 Fax: 858/752-6987 http://www.liai.org/pages/faculty-peters

— Reply to this email directly, view it on GitHub https://github.com/obi-ontology/obi/issues/1476#issuecomment-1115523970, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAAMMOOV3UZXL6GFBNGSECDVIB5LRANCNFSM5LUGYTDQ . You are receiving this because you were mentioned.Message ID: @.***>