Actions

Ontolog Forum

Revision as of 16:27, 1 June 2023 by Forum (talk | contribs) (→‎Resources)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Session COB
Duration 1 hour
Date/Time 25 Jan 2023 17:00 GMT
9:00am PST/12:00pm EST
5:00pm GMT/6:00pm CST
Convener Gary Berg-Cross

Ontology Summit 2023 COB

Helping scientific researchers make better use of ontologies

Agenda

  • 12:00 - 12:30 Chris Mungall The Core Ontology for Biology and Biomedicine (COB)
  • 12:30 - 13:00 Discussion

Conference Call Information

  • Date: Wednesday, 25 Jan 2023
  • Start Time: 9:00am PST / 12:00pm EST / 6:00pm CST / 5:00pm GMT / 1700 UTC
  • Expected Call Duration: 1 hour
  • Video Conference URL
    • Conference ID: 837 8041 8377
    • Passcode: 323309
  • Chat Room

The unabbreviated URLs are:

Attendees

Discussion

Welcome

RaviSharma: Welcome Everyone. Welcome to Chris Mungall.

Upper Model Requirements

Michael DeBellis: Perhaps this will be answered in the talk but has anyone defined the requirements and use cases for an Upper Model?

Sydney Cohen: If you're talking generally about top-level ontologies, there's this ISO/IEC standard - https://www.iso.org/standard/71954.html

Michael DeBellis: Thanks for the link Sydney

COB Requirements

Kai: Question from Damion Dooley and myself:

Should I assume that if I see a COB term and its COB id that originates from another OBO ontology term, that in adopting COB in an ontology, I can/should switch to the COB id? I understand that the SSSOM mapping file must indicate equivalency between the two.

And is this true too of properties? For example "is specified input of" is COB_0000027, and SSSOM file makes equivalency to OBI_0000295. I was puzzled because I thought COB would still stay away from minting common properties, instead leaving that to RO's domain.

James Overton: I'm sure Chris will talk about some of the things we need COB to do, but we do not have a formalized set of requirements and use cases for COB.

James Overton: This should (partially?) address Kai and Damion's question: http://obofoundry.org/COB/obo-bridge/

ToddSchneider: Do you develop requirements when there is disagreements about representations?

James Overton: Collecting requirements is a good approach to building community consensus. I would say that we're doing that more and more as try to align on design patterns.

Gary BC: Bio has some deceptively simple granular levels from systems, to organs to cells to molecules. But there may be an interesting connection to the more complex material/materials that Torsten mentioned.

James Overton: Keep in mind that the OBO community works at different scales. Projects may work in different ways, including gatherings fprmal requirements, and may have a long history of doing so. Strong coordination at the OBO Community level is relatively more recent, and driven by improvements to our tools and infastructure.

Gary BC: Torsten and others have found value in using common logic, rather than the more restritic DL/OWL for some of their models.

TorstenHahmann: I agree with Gary, sometimes it is easier to use first-order logic or Common Logic to delay dealing with some of the engineering problems

BFO Version

ToddSchneider: Which version of BFO is being used currently? The ISO version?

James Overton: RO uses a subset of BFO classes from BFO 1.1, defined here: https://github.com/oborel/obo-relations/blob/master/src/ontology/bfo-classes-minimal.owl

Independent Continuant

RaviSharma: what would be example of independant continuant?

James Overton: BFO 'independent continuant' http://purl.obolibrary.org/obo/BFO_0000004 That link should show a number of examples.

RaviSharma: I saw examples of independent continuant but at high level as process!

James Overton: 'independent continuant' example of usage: an organism; a leg; a molecule; the bottom right portion of a human torso; a heart; an atom; a chair; the interior of your mouth; a spatial region; an orchestra.

SKOS

Michael DeBellis: Have you considered using SKOS for some of the mapping? E.g., the two definitions of atom may be a skos:closeMatch.. almost synonyms but not quite

James Overton: Yes, skos:closeMatch is used for some of the links in that obo-bridge document.

has participant vs occurs in

John Judkins: When should we use 'has participant' and when should we use 'occurs in' to link a continuant to a process?

James Overton: @John: Roughly, P 'occurs in' C requires that the process P takes place completely "inside" the continuant C, while P 'has participant' C does not require that.

Schulz Conflation Model

Michael DeBellis: Does anyone have alink to the Schulz conflation model paper?

James Overton: The Stefan Schulz paper Chris mentioned https://jbiomedsem.biomedcentral.com/articles/10.1186/2041-1480-2-S2-S6

Michael DeBellis: Thanks James

Qualities

ToddSchneider: Are Qualities related to Units of Measure?

ToddSchneider: Was an ontology of Qualities developed?

Alignment with COB

Nico: A question I keep being asked: how do we “align with COB”? Practically.

Nico: So basically, instead of importing BFO directly we import COB! Thanks Chris

Chris Mungall: I agree.

Kai: I also agree.

Nico: But there are reservations on this, and it’s hard to coordinate

Nico: It basically means we have to allow cob classes without fixed bfo alignment!

Nico: Relevant: OBO Foundry is debating to require COB alignment moving forward!

Phenotype/Disease Integration into COB

RaviSharma: Are phenotype parameters totally independent of genetic changes?

Petr Kremen: Are there any plans/roadmap for the phenotype/disease integration into COB? Its about agreeing to Schulz conflation. If we can, we can do phenotype disease in less than a day!

Michael DeBellis: In general Phenotype = Genotype + Environment

Mike Bennett: Presumably in most cases a phenotype arises through the expression of some combination of genes interacting i.e. emergent?

So that would make Phenotype a Generically Dependent Continuant in BFO?

Alexander Diehl: The word phenotype is a conflation of descriptions of material entities and descriptions of processes and how these things may vary

I have argued for treating phenotypes as information artifacts that are about an organism, though the phenotype ontology folks generally reject this approach

Petr Kremen: Yeah, I expect st. like that - I was more curious about whether there is any driving force/project/activity that tries to drive it

Nico: The uPheno effort is pushing for it, but we could do with pressure

Petr Kremen: Thanks Nico and Chris for info, will check.

Acknowledgments

Michael DeBellis: Thanks Chris, really interesting

Tom Walpole: Good Stuff - Thank You

Petr Kremen: Thanks Chris, great job!

Ravi Sharma: Thanks chris and torsten and James and Gary and Ken.

Nico: Awesome talk thanks Chris!

Paul Fabry: Great talk. Thank you!

Jacques Hilbey: Thanks!

Resources

Previous Meetings

 Session
ConferenceCall 2023 01 18Launch
ConferenceCall 2022 12 14Planning
ConferenceCall 2022 12 07Data Principles
... further results

Next Meetings

 Session
ConferenceCall 2023 02 01ROBOT
ConferenceCall 2023 02 08ODK
ConferenceCall 2023 02 15OBO Dashboards
... further results