Review allow_equivalents: all design decision for BERO

[hrshdhgd]

to avoid having to hack OBI, ENVO and NCIT:

ERROR Equivalence: <http://purl.obolibrary.org/obo/NCIT_C179199> == <http://purl.obolibrary.org/obo/NCIT_C1127>
ERROR Equivalence: <http://purl.obolibrary.org/obo/ENVO_01001479> == <http://purl.obolibrary.org/obo/ENVO_01001784>
ERROR Equivalence: <http://purl.obolibrary.org/obo/OBI_2100024> == <http://purl.obolibrary.org/obo/OBI_0003012>
ERROR Equivalence: <http://purl.obolibrary.org/obo/ENVO_01001483> == <http://purl.obolibrary.org/obo/ENVO_00010504>
ERROR Equivalence: <http://purl.obolibrary.org/obo/OBI_0002781> == <http://purl.obolibrary.org/obo/OBI_0002783>

This occurs during the prepare_release step :

sh run.sh make prepare_release IMP=false

We decided to skip the checks for inferred equivalences. Alternatively we could remove the offending axioms from the ontologies 1 by 1.

Thoughts @cmungall ?

[cmungall]

You are doing the right thing in skipping for now But we should simultaneously look into these do you know you to use robot to generate explanations for these?

…

On Thu, May 26, 2022 at 3:22 AM Harshad ***@***.***> wrote: to avoid having to hack OBI, ENVO and NCIT: ERROR Equivalence: <http://purl.obolibrary.org/obo/NCIT_C179199> == <http://purl.obolibrary.org/obo/NCIT_C1127> ERROR Equivalence: <http://purl.obolibrary.org/obo/ENVO_01001479> == <http://purl.obolibrary.org/obo/ENVO_01001784> ERROR Equivalence: <http://purl.obolibrary.org/obo/OBI_2100024> == <http://purl.obolibrary.org/obo/OBI_0003012> ERROR Equivalence: <http://purl.obolibrary.org/obo/ENVO_01001483> == <http://purl.obolibrary.org/obo/ENVO_00010504> ERROR Equivalence: <http://purl.obolibrary.org/obo/OBI_0002781> == <http://purl.obolibrary.org/obo/OBI_0002783> This occurs during the prepare_release step : sh run.sh make prepare_release IMP=false We decided to skip the checks for inferred equivalences. Alternatively we could remove the offending axioms from the ontologies 1 by 1. Thoughts @cmungall <https://github.com/cmungall> ? — Reply to this email directly, view it on GitHub <#3>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAAMMOLABHL7V6FYDXDLNF3VL5GF5ANCNFSM5XAXR5XQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[hrshdhgd]

Do you mean under robot_report as seen here ?

[matentzn]

I have tickets in for most of these issues for a while.

• Accidental equivalent definitions EnvironmentOntology/envo#1069 (check my last comments)
• Add QC check to prevent inferred equivalencies NCI-Thesaurus/thesaurus-obo-edition#76

[matentzn]

This is an example how you get explanations:

sh run.sh robot explain -i ../../bero.owl --axiom "'Omacetaxine' EquivalentTo 'Omacetaxine Mepesuccinate'" -o explain.md

Can you also share the output of ls -l imports/*_import.owl to see how large the imports are?

[hrshdhgd]

explain.md: It says "No explanations found."

Just for documentation purposes, the correct command is

sh run.sh robot explain -i ../../bero.owl --axiom "'Omacetaxine' EquivalentTo 'Omacetaxine Mepesuccinate'" --explanation explain.md

-o => --explanation

Imported files:

-rw-r--r--  1 HHegde  staff  504065193 May 26 13:10 src/ontology/imports/chebi_import.owl
-rw-r--r--  1 HHegde  staff     147574 May 26 12:58 src/ontology/imports/ecocore_import.owl
-rw-r--r--  1 HHegde  staff     147857 May 26 13:27 src/ontology/imports/edam_import.owl
-rw-r--r--  1 HHegde  staff    6583353 May 26 13:11 src/ontology/imports/envo_import.owl
-rw-r--r--  1 HHegde  staff  137010643 May 26 13:13 src/ontology/imports/go_import.owl
-rw-r--r--  1 HHegde  staff   11500194 May 26 13:28 src/ontology/imports/mesh_import.owl
-rw-r--r--  1 HHegde  staff    3760208 May 26 12:57 src/ontology/imports/mop_import.owl
-rw-r--r--  1 HHegde  staff   18048921 May 26 13:10 src/ontology/imports/ncbitaxon_import.owl
-rw-r--r--  1 HHegde  staff  344918918 May 26 13:27 src/ontology/imports/ncit_import.owl
-rw-r--r--  1 HHegde  staff    7701400 May 26 12:58 src/ontology/imports/obi_import.owl
-rw-r--r--  1 HHegde  staff      76487 May 26 12:58 src/ontology/imports/omo_import.owl
-rw-r--r--  1 HHegde  staff     676248 May 26 13:27 src/ontology/imports/pato_import.owl
-rw-r--r--  1 HHegde  staff    5853306 May 26 12:58 src/ontology/imports/po_import.owl
-rw-r--r--  1 HHegde  staff   12063663 May 26 12:59 src/ontology/imports/to_import.owl

Released files:

-rw-r--r--  1 HHegde  staff  1080249848 May 26 14:40 ../../bero.owl
-rw-r--r--  1 HHegde  staff   425830749 May 26 14:40 ../../bero.obo
-rw-r--r--  1 HHegde  staff  1001991731 May 26 14:40 ../../bero.json
-rw-r--r--  1 HHegde  staff  1083874906 May 26 14:39 ../../bero-full.owl
-rw-r--r--  1 HHegde  staff   425830778 May 26 14:39 ../../bero-full.obo
-rw-r--r--  1 HHegde  staff  1001991746 May 26 14:39 ../../bero-full.json
-rw-r--r--  1 HHegde  staff        2547 May 26 14:39 ../../bero-base.owl
-rw-r--r--  1 HHegde  staff         537 May 26 14:39 ../../bero-base.obo
-rw-r--r--  1 HHegde  staff        1695 May 26 14:39 ../../bero-base.json

[matentzn]

Explain: check here http://robot.obolibrary.org/explain

Instead of -o write --explanation

[matentzn]

I have extended the pipeline to generate all related explanations:

https://github.com/berkeleybop/bero/blob/main/src/ontology/bero.Makefile#L95

Something is wrong though:

reason --equivalent-classes-allowed asserted-only results in the output in the first comment of this post;

but NONE of the explanations work reveal anything.

@balhoff do you have an idea of why, for example:

https://github.com/berkeleybop/bero/blob/main/src/ontology/bero.Makefile#L87

reveals no explanation, while the reasoner reveals this exact equivalence as inferred during reason?

[balhoff]

@matentzn sometimes ELK has trouble giving explanations when it's the result of mutual subClassOf relations. Could that be the case here?

[matentzn]

defin. possible! Thank you!

[matentzn]

@hrshdhgd in this case, i dont know how to debug this ticket without major efforts. We will also not learn why NCI-Thesaurus/thesaurus-obo-edition#76 which is a pity, since @cmungall cannot reproduce it.

[cmungall]

https://douroucouli.wordpress.com/2018/09/04/debugging-ontologies-using-owl-reasoning-part-2-unintentional-entailed-equivalence/

…

On Fri, May 27, 2022 at 5:47 AM Nico Matentzoglu ***@***.***> wrote: @hrshdhgd <https://github.com/hrshdhgd> in this case, i dont know how to debug this ticket without major efforts. We will also not learn why NCI-Thesaurus/thesaurus-obo-edition#76 <NCI-Thesaurus/thesaurus-obo-edition#76> which is a pity, since @cmungall <https://github.com/cmungall> cannot reproduce it. — Reply to this email directly, view it on GitHub <#3 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAAMMONODEQ62BXMGEI25XLVMC77PANCNFSM5XAXR5XQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[matentzn]

Hahah yes. Good luck @hrshdhgd :P