Proposed cleanup of specimen hierarchy

[sebastianduesing]

Closes #1821. This pull request is a reorganization & cleanup of the specimen hierarchy to fix issues identified during the COB workshop. Changes are as follows:

• Some terms from the specimen branch that should have been in the template were not. This PR moves all subclasses of "specimen" into the specimen template.
• Unlike most other templates, the specimen template did not have a dedicated column for parent class. Instead, all of its axiom columns were formatted as "specimen and _______". This PR removes the "specimen and" part of those columns and adds a column for parent class.
• This PR makes all terms for specimens from an organism asserted subclasses of "specimen from organism".
• Some logic changes were made during the process of moving terms into the template; these are primarily the result of translating unnecessarily complicated or redundant axioms into the standard template forms. I've done my best to ensure the template axiomatization captures what the original axiomatization was intended to capture.
• An axiom adapted from the process "collecting specimen from organism" was added to "specimen from organism" to facilitate logical linkage between those two terms & terms referencing either of them.
• I made no other changes to specimen terms when moving them into the templates except for a limited number of cases of specimens referencing anatomical parts from Uberon. OBI has many specimen terms that follow a pattern of "{anatomical part} specimen", all of which have similar axioms referencing the anatomical part from which the specimen was derived. There were a small number of terms that had no such axiom despite it being clear which anatomical part they were intended to reference in such an axiom. In these cases, I added the appropriate axiom. Most used terms already imported into OBI that just were not referenced in the term's logic. For one, "cloacal specimen", the term "cloaca" was not already imported, so I imported it.

[jamesaoverton]

Discussed on the OBI developer call 2024-10-28: Looks good to us. @jamesaoverton will do a technical review before merging.

[jamesaoverton]

@sebastianduesing Please add a check like https://github.com/obi-ontology/obi/blob/master/src/sparql/assay-violation-modules.rq for the specimen branch.

[sebastianduesing]

It looks like this check is catching terms that reason under specimen, not just those that are asserted subclasses of specimen, e.g., 'material sample' and 'nucleic acid extract' and their subclasses. Should those be added to the template as well?

Adjacent to that, it's a bit unclear to me what the intended distinction between 'material sample' and 'specimen' actually is.

[bpeters42]

- Are we trying to clean up all of specimens, or just 'specimen from organism'? I would start narrower, and worry about the others later. - A sample is collected with an intent for it to be representative of a larger whole, e.g. a tube of water collected from a lake to monitor oxygen levels in the lake. There are specimens collected for which there is no obvious 'whole', like Einstein's brain in a jar, or a meteorite.

…

On Mon, Nov 4, 2024 at 11:11 AM Sebastian Duesing ***@***.***> wrote: It looks like this check is catching terms that reason under specimen, not just those that are asserted subclasses of specimen, e.g., 'material sample' and 'nucleic acid extract' and their subclasses. Should those be added to the template as well? Adjacent to that, it's a bit unclear to me what the intended distinction between 'material sample' and 'specimen' actually is. — Reply to this email directly, view it on GitHub <#1830 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADJX2IUSNOXD3LPBZQI6YNLZ662FFAVCNFSM6AAAAABPVM2QLSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDINJVGQ4TSMJSGM> . 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

[sebastianduesing]

It is my understanding that we wanted to clean up the whole specimen branch. The gist of my previous question is whether "the whole specimen branch" includes only asserted descendants of 'specimen' or all classes that infer under 'specimen'. E.g., 'nucleic acid extract' (one of the terms being flagged by this check) is defined as a subclass of 'extract', but it reasons under 'processed specimen'. It would not be hard or time-consuming to add these other classes to the specimen template, if that's what I should do.

Thanks for the answer about the distinction between sample and specimen. I think I'd missed the part/whole thing. That makes sense. Sample is not currently a subclass of specimen, but I am thinking that it ought to be.

[jamesaoverton]

Thanks @sebastianduesing. I wasn't expecting those 15 terms to fail the new test, but I think it's good that they did. All of them clearly fit under 'specimen', and I would like to have all specimens in a template, like we do for assays, devices, study designs, and several other important branches of OBI. So I would ask you to do two things to finish that work in this PR:

1. move the 15 terms that fail that new test out of obi-edit.owl and into the specimen template: 'material sample', 'whole organism preparation', 'nucleic acid extract', and their descendants
2. improve a couple of logical definitions, which will improve the asserted hierarchy

• 'whole organism preparation' replace subClassOf "'material entity' and ('has role' some 'specimen role')" with subClassOf 'specimen' (which means the same thing)
• 'nucleic acid extract' replace subClassOf "(is_specified_output_of some 'nucleic acid extraction') and ('has role' some 'specimen role')" with two simpler axions: subClassOf "(is_specified_output_of some 'nucleic acid extraction')" and subClassOf specimen
• 'material sample' replace equivalentTo "'material entity' and ('has role' some 'material sample role')" with "specimen and ('has role' some 'material sample role')"

[sebastianduesing]

Thanks @jamesaoverton, that all makes sense to me. I'll do that.

[jamesaoverton]

Thanks!