Actions

ConferenceCall 2023 02 08: Difference between revisions

Ontolog Forum

Tag: Reverted
 
(10 intermediate revisions by the same user not shown)
Line 5: Line 5:
|-
|-
! scope="row" | Duration
! scope="row" | Duration
| [[duration::1 hour]]
| [[duration::1.5 hour]]
|-
|-
! scope="row" rowspan="3" | Date/Time
! scope="row" rowspan="3" | Date/Time
Line 15: Line 15:
|-
|-
! scope="row" | Convener
! scope="row" | Convener
| [[convener::GaryBergCross|Gary Berg-Cross]]
| [[convener::JamesOverton|James Overton]]
|}
|}


Line 23: Line 23:
== Agenda ==
== Agenda ==
* The Ontology Development Kit (ODK)
* The Ontology Development Kit (ODK)
* Presenters: Nico Matentzoglu, Damien Goutte-Gattat, Philip Stroemert, Ray Stefancsik andor Anita Caron
* Presenters: Nico Matentzoglu, Damien Goutte-Gattat, Philip Strömert, Ray Stefancsik and/or Anita Caron
* [https://bit.ly/40MyKsj Slides]
* [https://bit.ly/3Xo8ku6 Setting up your own repository]
* [https://bit.ly/3YDzVIJ Video Recording]


== Conference Call Information ==
== Conference Call Information ==
Line 29: Line 32:
* Start Time: 9:00am PST / 12:00pm EST / 6:00pm CST / 5:00pm GMT / 1700 UTC  
* Start Time: 9:00am PST / 12:00pm EST / 6:00pm CST / 5:00pm GMT / 1700 UTC  
** ref: [http://www.timeanddate.com/worldclock/fixedtime.html?month=02&day=08&year=2023&hour=12&min=00&sec=0&p1=179 World Clock]  
** ref: [http://www.timeanddate.com/worldclock/fixedtime.html?month=02&day=08&year=2023&hour=12&min=00&sec=0&p1=179 World Clock]  
* Expected Call Duration: 1 hour
* Expected Call Duration: 1.5 hour
{{:OntologySummit2023/ConferenceCallInformation}}
{{:OntologySummit2023/ConferenceCallInformation}}


== Attendees ==
== Attendees ==
* [[AleixPuig|Aleix Puig]]
* [[AlexanderDiehl|Alexander Diehl]]
* [[AlexShkotin|Alex Shkotin]]
* [[AllenBaron|Allen Baron]]
* [[AndreaWesterinen|Andrea Westerinen]]
* [[AnitaCaron|Anita Caron]]
* [[ArwaIbrahim|Arwa Ibrahim]]
* [[BartGajderowicz|Bart Gajderowicz]]
* [[BobbinTeegarden|Bobbin Teegarden]]
* [[ClaudiaSanchezBeatoJohnson|Claudia Sanchez-Beato]]
* [[CMAlvarez|C.M. Alvarez]]
* [[DamianGoutteGattat|Damian Goutte-Gattat]]
* [[FanLi|Fan Li]]
* [[GaryBergCross|Gary Berg-Cross]]
* [[JamesOverton|James Overton]]
* [[JenniferGiron|Jennifer Girón]]
* [[JieZheng|Jie Zheng]]
* [[JimBalhoff|Jim Balhoff]]
* [[JinjingGuo|Jinjing Guo]]
* [[KenBaclawski|Ken Baclawski]]
* [[MikeBennett|Mike Bennett]]
* [[NancyWiegand|Nancy Wiegand]]
* [[NicoMatentzoglu|Nico Matentzoglu]]
* [[PhilippeRoccaSerra|Philippe Rocca-Serra]]
* [[PhilipStroemert|Philip Strömert]]
* randi
* [[RaviSharma|Ravi Sharma]]
* [[RayStefancsik|Ray Stefancsik]]
* [[RobertRovetto|Robert Rovetto]]
* [[SabrinaToro|Sabrina Toro]]
* [[StefanoDeGiorgis|Stefano De Giorgis]]
* [[SydneyCohen|Sydney Cohen]]
* [[UgurBayindir|Ugur Bayindir]]


== Discussion ==
== Discussion ==
[12:06] RaviSharma: James Welcome
[11:56] RaviSharma: welcome everyone


[12:06] James Overton: ROBOT slides: http://james.overton.ca/slides/robot-tutorial-2023/index.html#/title-slide
[12:07] RaviSharma: Nico, So we find common word in two or more ontologies but do not we have to also see that content meaning of blood constituents are also the same?


OAK slides: https://docs.google.com/presentation/d/1XK2fOlfLVXc25DrVFfS3kpt5jBpX8fQ3OePsM5rlSNE/edit#slide=id.p
[12:10] RaviSharma: Nico, we like the common infrastructure with tools that gives stability like the Operating systems did in early computing


[12:08] RaviSharma: Philosophy is mother of Mathematics and Physics or for that matter logic and reasoning. We sometimes call it metaphysics also. We are proud of Philosophers, and welcome you. This also has basis of ontology.
[12:11] James Overton: Ravi: The import system is built on identifiers (CURIEs/IRIs) not labels. So we would ask for UBERON:0000178 'blood'


[12:15] RaviSharma: James can you explain the power of Query a bit more?
[12:17] RaviSharma: Damien, does github provide or your team has a regular compatibility check or do you execute scripts for Make to test integrability?


[12:16] RaviSharma: James does ROBOT work with Franz Gruff?
[12:20] James Overton: Ravi: ODK uses GitHub Actions for continuous integration testing, which calls Make tests: https://github.com/INCATools/ontology-development-kit/tree/master/.github/workflows


[12:17] RaviSharma: James, what are the commands relevant to construct KG?
[12:27] RaviSharma: Damien - are the output files run by a user tagged so that there is versioning by users running similar commands?


[12:19] RaviSharma: James, how do we use some commands for NLP?
[12:28] RaviSharma: how different commands are available and is there a ? option!


[12:22] RaviSharma: James you have given users power of Query I see now.
[12:30] Allen Baron: Great question and very clear explanation. Thanks Damien.


[12:23] RaviSharma: James, if you were or if community were to add commands, which ones would you prioritize?
[12:32] From  Nico Matentzoglu: Damien, maybe we could run “./run.sh ls /tools/“ to get a good set of most apps, and also “./run.sh pip list” for the python stuff, and we could perhaps curate a `/tools/help` script prints these..


[12:29] Nico Matentzoglu: What is Franz Gruff?
[12:32] James Overton: Ravi: ODK does not tag files. We rely on git for version control, and branch-and-merge workflows, usually supported by GitHub or GitLab.


Added by Ken Baclawski: Gruff is a free Knowledge Graph visualization product from Franz Inc. https://allegrograph.com/products/gruff/
[12:33] James Overton: Damien's answer for Ravi: The ODK paper includes a list of all the included tools, but it would be good to add a documentation page with a list of common ODK tools/commands.


[12:29] Nico Matentzoglu: No commands for NLP directly in ROBOT, no
[12:35] Damien Goutte-Gattat (FlyBase): `help` would conflict with the shell builtin of the same name. But overall, yeah, some kind of inline help would be good.


[12:32] Nico Matentzoglu: Ravi, re prioritizing commands, we do that here: https://github.com/ontodev/robot/issues It depends on the major projects at the time
[12:44] David Osumi-Sutherland: Damien, maybe we could


[12:32] Nico Matentzoglu: But its a good question to ask in the QA later


[12:32] RaviSharma: James great tutorial a bit rushed at end but great tool and lot of capabilities, thanks.
[12:40] RaviSharma: Philip - what are the minimum metadata beside ID of ontology so the rest are taken as defaults, as many users may not have knowledge of ODKfull?


[12:33] C.M. Alvarez: Do you have the slack link?
[12:45] RaviSharma: Philip great comment about refresh version, but how do you handle compatibility between older versions of out integration namely provenance compatibility?


[12:33] RaviSharma: Chris Welcome back
[12:51] RaviSharma: For Anita or any presenters - do you have places where you store error files?


[12:34] James Overton: Thanks Ravi. Yes, these slides were written for a longer presentation slot. I encourage anyone to look at the documentation site, or go over my slides more slowly: http://james.overton.ca/slides/robot-tutorial-2023/index.html
[12:55] Sydney Cohen: is there anywhere to find a more detailed report for the violations of the SPARQL checks?


[12:35] RaviSharma: Chris, does OAK work with other than english language, same Q for James?
[12:55] Sydney Cohen: would that go into the reports folder?


[12:35] Michael DeBellis: Chrome gives me a privacy
[12:56] James Overton: The built-in ROBOT report commands each have a page with "problem-solution" documentation: http://robot.obolibrary.org/report_queries/'


[12:36] Nico Matentzoglu: Me too. it's because you are going into a non-https side.. I didn't have that last time.
[12:56] James Overton: And ROBOT reports can output in text, TSV, and HTML formats.


[12:36] Gary Berg-Cross: Yes, MD, but you can proceed safely
[12:57] James Overton: Small HTML example: http://dashboard.obofoundry.org/dashboard/obi/robot_report.html


[12:36] Michael DeBellis: How does Oak compare with OWLReady2?
[12:57] Sydney Cohen: okay thank you


[12:38] James Overton: Ravi: Yes, but we could do better. We support RDF and OWL, which use Unicode and language tags, but we could do better to support multiple labels in different languages. I would *love* to do better, but I haven't had a project to fund that work.
[13:01] Ugur Bayindir: Side question; the editor Anita was using looks neat, what was that? And do you suggest using it in daily ontology editing tasks?


[12:39] RaviSharma: Ken, can you kindly comment on KGSQL if it is applicable or interoperable?
[13:03] Anita Caron: I agree.


Added by Ken Baclawski: KGSQL is applicable.  It extends KGs with additional features that would be useful.  However, KGSQL is not yet generally supported so it is not interoperable.
[13:01] Allen Baron: Looked like Microsoft VS code to me.


[12:39] James Overton: While there's some overlap, I would say that OAK is higher-level and more focused on browsing and analysis than OwlReady2.
[13:03] Anita Caron: I agree.


[12:40] RaviSharma: Mike Bennett - congrats on being a technical director at OMG. Very proud of your accomplishments.
[13:01] Allen Baron: …. and Protege for editing the ontology file itself.


[12:40] Mike Bennett: Thanks Ravi!
[13:08] Anita Caron: I agree.


[12:41] Mike Bennett: Great to see an ontology viz that uses a line for a restriction instead of a Nameless Box. Critical for business SME validation!
[13:02] Anita Caron: Hi Ugur, yes, I recommend using VS code in your daily ontology editing.


[12:42] RaviSharma: Chris, as compared to older GO tools how are we using OAK or ROBOT
[13:03] Ugur Bayindir: I agree.


[12:46] Vincent Vialard: Did you try the Konclude reasoner for Robot? It is both fast and complete (but not in Java)...
[13:02] Mike Bennett: My concern is that in such a deep dev environment, devs may write tests that only apply to application-focused ontologies (e.g. we saw redundant sub-class relations) which may not always be applicable to a concept ontology. Process needs to account for tests v stated style / aim of the ontology itself.


[12:47] James Overton: No, we haven't tried Konclude.
[13:04] Nico Matentzoglu: Very good point Mike, I totally agree


[12:49] Mike Bennett: An idea: would it be possible to use Robot or OAK to derive operational ontologies from a purely conceptual ontology (i.e. inserting datatypes and data surrogates for real world things)?
[13:05] Nico Matentzoglu: Customisability is key±


[12:50] StroemPhi: https://owlready2.readthedocs.io/en/v0.37/
[13:08] Allen Baron: What’s included the documentation ODK can automatically generate about an ontology?


[12:51] Chris Mungall: https://github.com/INCATools/ontology-access-kit/issues/102 - add links to other tools here!
[13:11] Anita Caron: Complete tutorial for adding custom SPARQL check https://oboacademy.github.io/obook/tutorial/custom-qc/


[12:52] RaviSharma: Chris and James - appreciate your answering many or most of my Qs.
[13:12] C.M. Alvarez (USBP): I agree.


[12:53] RaviSharma: What XML capabilities are built in these commands, mostly for transport?
[13:02] RaviSharma: James and Team today - great integration development and output effort.


[12:54] James Overton: LinkML data modelling tool: http://linkml.io/
[13:05] James Overton: Yes, the editor Anita was using is VS Code.


[12:59] RaviSharma: Again many thanks for two great presentations.
[13:11] RaviSharma: Thanks for informing that 40-70 users use this toolset and apps regularly.


[12:59] C.M. Alvarez: Thanks for a great presentation!
[13:19] James Overton: 40-70 ontology projects use ODK, that we know of. It's an open source project, so who knows how many people use it.
 
[13:02] Thomas Exner: Thanks for the presentations. What were the other things you mentioned similar to LinkML?
 
[13:03] James Overton: LinkML is similar to various validation/schema languages such as JSON Schema, ShEx, and SHACL.
 
[13:03] Thomas Exner: Thanks.
 
[13:13] James Overton: XML capabilities: We use RDF/XML as our default data format, but support other RDF and OWL formats.


== Resources ==
== Resources ==
* [https://bit.ly/3YDzVIJ Video Recording]
* [https://oboacademy.github.io/obook/tutorial/custom-qc/ Tutorial: How to add custom quality checks with ODK]
* [https://youtu.be/HAprZl0pldk Video Recording on YouTube]


== Previous Meetings ==
== Previous Meetings ==

Latest revision as of 16:28, 1 June 2023

Session ODK
Duration 1.5 hour
Date/Time 08 Feb 2023 17:00 GMT
9:00am PST/12:00pm EST
5:00pm GMT/6:00pm CST
Convener James Overton

Ontology Summit 2023 ODK

Helping scientific researchers make better use of ontologies

Agenda

Conference Call Information

  • Date: Wednesday, 08 Feb 2023
  • Start Time: 9:00am PST / 12:00pm EST / 6:00pm CST / 5:00pm GMT / 1700 UTC
  • Expected Call Duration: 1.5 hour
  • Video Conference URL
    • Conference ID: 837 8041 8377
    • Passcode: 323309
  • Chat Room

The unabbreviated URLs are:

Attendees

Discussion

[11:56] RaviSharma: welcome everyone

[12:07] RaviSharma: Nico, So we find common word in two or more ontologies but do not we have to also see that content meaning of blood constituents are also the same?

[12:10] RaviSharma: Nico, we like the common infrastructure with tools that gives stability like the Operating systems did in early computing

[12:11] James Overton: Ravi: The import system is built on identifiers (CURIEs/IRIs) not labels. So we would ask for UBERON:0000178 'blood'

[12:17] RaviSharma: Damien, does github provide or your team has a regular compatibility check or do you execute scripts for Make to test integrability?

[12:20] James Overton: Ravi: ODK uses GitHub Actions for continuous integration testing, which calls Make tests: https://github.com/INCATools/ontology-development-kit/tree/master/.github/workflows

[12:27] RaviSharma: Damien - are the output files run by a user tagged so that there is versioning by users running similar commands?

[12:28] RaviSharma: how different commands are available and is there a ? option!

[12:30] Allen Baron: Great question and very clear explanation. Thanks Damien.

[12:32] From Nico Matentzoglu: Damien, maybe we could run “./run.sh ls /tools/“ to get a good set of most apps, and also “./run.sh pip list” for the python stuff, and we could perhaps curate a `/tools/help` script prints these..

[12:32] James Overton: Ravi: ODK does not tag files. We rely on git for version control, and branch-and-merge workflows, usually supported by GitHub or GitLab.

[12:33] James Overton: Damien's answer for Ravi: The ODK paper includes a list of all the included tools, but it would be good to add a documentation page with a list of common ODK tools/commands.

[12:35] Damien Goutte-Gattat (FlyBase): `help` would conflict with the shell builtin of the same name. But overall, yeah, some kind of inline help would be good.

[12:44] David Osumi-Sutherland: Damien, maybe we could


[12:40] RaviSharma: Philip - what are the minimum metadata beside ID of ontology so the rest are taken as defaults, as many users may not have knowledge of ODKfull?

[12:45] RaviSharma: Philip great comment about refresh version, but how do you handle compatibility between older versions of out integration namely provenance compatibility?

[12:51] RaviSharma: For Anita or any presenters - do you have places where you store error files?

[12:55] Sydney Cohen: is there anywhere to find a more detailed report for the violations of the SPARQL checks?

[12:55] Sydney Cohen: would that go into the reports folder?

[12:56] James Overton: The built-in ROBOT report commands each have a page with "problem-solution" documentation: http://robot.obolibrary.org/report_queries/'

[12:56] James Overton: And ROBOT reports can output in text, TSV, and HTML formats.

[12:57] James Overton: Small HTML example: http://dashboard.obofoundry.org/dashboard/obi/robot_report.html

[12:57] Sydney Cohen: okay thank you

[13:01] Ugur Bayindir: Side question; the editor Anita was using looks neat, what was that? And do you suggest using it in daily ontology editing tasks?

[13:03] Anita Caron: I agree.

[13:01] Allen Baron: Looked like Microsoft VS code to me.

[13:03] Anita Caron: I agree.

[13:01] Allen Baron: …. and Protege for editing the ontology file itself.

[13:08] Anita Caron: I agree.

[13:02] Anita Caron: Hi Ugur, yes, I recommend using VS code in your daily ontology editing.

[13:03] Ugur Bayindir: I agree.

[13:02] Mike Bennett: My concern is that in such a deep dev environment, devs may write tests that only apply to application-focused ontologies (e.g. we saw redundant sub-class relations) which may not always be applicable to a concept ontology. Process needs to account for tests v stated style / aim of the ontology itself.

[13:04] Nico Matentzoglu: Very good point Mike, I totally agree

[13:05] Nico Matentzoglu: Customisability is key±

[13:08] Allen Baron: What’s included the documentation ODK can automatically generate about an ontology?

[13:11] Anita Caron: Complete tutorial for adding custom SPARQL check https://oboacademy.github.io/obook/tutorial/custom-qc/

[13:12] C.M. Alvarez (USBP): I agree.

[13:02] RaviSharma: James and Team today - great integration development and output effort.

[13:05] James Overton: Yes, the editor Anita was using is VS Code.

[13:11] RaviSharma: Thanks for informing that 40-70 users use this toolset and apps regularly.

[13:19] James Overton: 40-70 ontology projects use ODK, that we know of. It's an open source project, so who knows how many people use it.

Resources

Previous Meetings

 Session
ConferenceCall 2023 02 01ROBOT
ConferenceCall 2023 01 25COB
ConferenceCall 2023 01 18Launch
... further results

Next Meetings

 Session
ConferenceCall 2023 02 15OBO Dashboards
ConferenceCall 2023 02 22Ubergraph
ConferenceCall 2023 03 01Panel
... further results
Retrieved from "https://ontologforum.com/index.php?title=ConferenceCall_2023_02_08&oldid=4601"