obophenotype
Add Axl+ dendritic cell
Related to issue #764.
I was not able to think about a logical definition, maybe @addiehl or @masciam have suggestions?
Here's a diff of how these changes impact the classified ontology (on -simple file):
Ontology comparison
Left
- Ontology IRI:
http://purl.obolibrary.org/obo/cl/cl-simple.owl - Version IRI:
http://purl.obolibrary.org/obo/cl/releases/2023-01-06/cl-simple.owl - Loaded from:
file:/__w/cell-ontology/cell-ontology/src/ontology/cl-simple-master.owl/cl-simple.owl
Right
- Ontology IRI:
http://purl.obolibrary.org/obo/cl/cl-simple.owl - Version IRI:
http://purl.obolibrary.org/obo/cl/releases/2023-01-06/cl-simple.owl - Loaded from:
file:/__w/cell-ontology/cell-ontology/src/ontology/cl-simple-pr.owl/cl-simple.owl
Ontology imports
Ontology annotations
CL_0017009 http://purl.obolibrary.org/obo/CL_0017009
Added
-
CL_0017009 label "Axl+ dendritic cell, human"
-
CL_0017009 definition "A human dendritic cell that expresses the AXL and SIGLEC6 genes."
-
database_cross_reference "PMID:29221729"
-
database_cross_reference "PMID:28428369"
-
database_cross_reference "PMID:31156637"
-
-
CL_0017009 has_exact_synonym "AS DC, human"
-
CL_0017009 date "2022-08-25T14:18:37Z"^^dateTime
-
Class: CL_0017009
-
CL_0017009 SubClassOf dendritic cell, human
![github-actions[bot] avatar](https://avatars.githubusercontent.com/in/15368?v=4 "Published by a pub.dev verified publisher"){kind=small}
Here's a diff of your edit file (unreasoned)
Ontology comparison
Left
- Ontology IRI:
http://purl.obolibrary.org/obo/cl.owl - Version IRI: None
- Loaded from:
file:/__w/cell-ontology/cell-ontology/master/src/ontology/cl-edit.owl
Right
- Ontology IRI:
http://purl.obolibrary.org/obo/cl.owl - Version IRI: None
- Loaded from:
file:/__w/cell-ontology/cell-ontology/branch/src/ontology/cl-edit.owl
Ontology imports
Ontology annotations
CL_0017009 http://purl.obolibrary.org/obo/CL_0017009
Added
-
CL_0017009 label "Axl+ dendritic cell, human"
-
CL_0017009 definition "A human dendritic cell that expresses the AXL and SIGLEC6 genes."
-
database_cross_reference "PMID:31156637"^^string
-
database_cross_reference "PMID:28428369"^^string
-
database_cross_reference "PMID:29221729"^^string
-
-
CL_0017009 has_exact_synonym "AS DC, human"
-
CL_0017009 date "2022-08-25T14:18:37Z"^^dateTime
-
Class: CL_0017009
-
CL_0017009 SubClassOf expresses some sialic acid-binding Ig-like lectin 6 (human)
-
CL_0017009 SubClassOf dendritic cell, human
-
CL_0017009 SubClassOf expresses some tyrosine-protein kinase receptor UFO (human)
2 things need to be added:
- [x] in_taxon some homo_sapien
- [ ] expresses AXL, expresses SIGLEC6 <- CL uses PRO atm despite inaccuracy of protein vs gene (might be related to this PR https://github.com/obophenotype/cell-ontology/pull/901)
@shawntanzk in_taxon some homo_sapien is inferred from subclassOf "dendritic cell, human", isn't it?
This PR has not seen any activity in the past month; if nobody comments or reviews it in the next week, the PR editor will be allowed to proceed with merging without explicit approval, should they wish to do so.
@lubianat, checking in on this PR. It has no reviewers and seems to have no activity for several months. Can you merge or close?
@bvarner-ebi thanks! @shawntanzk pointed I should add some way of saying the cell expresses the genes. How should I do that, though?
I think both proteins are present in the plasma membrane at some point, but is there a more general way that does not depend on the protein in the plasma membrane?
@bvarner-ebi thanks! @shawntanzk pointed I should add some way of saying the cell expresses the genes. How should I do that, though?
expresses some PR:000001928
expresses some PR:000004528
Let me know if you need help adding those in the PR obo slim :)
@shawntanzk I don't know what you mean by "OBO slim", I'm sorry. By PR you mean Protein Ontology or Pull Request?
@shawntanzk I don't know what you mean by "OBO slim", I'm sorry
ah yeah, we really need to document this. Basically PR ontology is too big, so we have a obo slim version where we add terms we care about: https://github.com/obophenotype/pro_obo_slim We import that instead of the whole PR, so new terms might have to be added there before we import them into CL. if this isn't urgent, give me some time to write docs carpal tunnel is killing me a bit atm, so writing full docs sounds painful >.<
@shawntanzk Not urgent for me; take care of your carpal tunnel! It looks a bit cumbersome, so I'll wait for the docs. Thanks!
@lubianat, it's been a while since I did this, but here are my notes on adding PRO terms to CL after @shawntanzk helped me do this. If you try this, can you let us know how it goes? If successful, these notes can be added to documentation...
- Add desired PRO term to seed.txt in pro_obo_slim
- Open docker
- Open terminal
- Navigate (
cd) to the main directory in pro_obo_slim - Run
sh odk.sh make all - Make a pull request and merge all updates - can merge this without review
- In the cell-ontology repo, create a branch
- Add PRO term to src/ontology/imports/pr_terms.txt
- Go to terminal, navigate to cell-ontology/src/ontology
- Run
sh run.sh make mirror/pr.owl -b(Note:-bforces command to run even if the mirror is up-to-date) - Run
sh run.sh make imports/merged_import.owl - Open Protégé. The PRO term should now be available for modelling END
@Thank you bvarner!
We should have a simpler way to automate all of that...
My step-by-step:
After having Docker installed and configured:
- Clone https://github.com/obophenotype/pro_obo_slim.
- Edit locally https://github.com/obophenotype/pro_obo_slim/blob/master/seed.txt to add the Protein Ontology terms desired.
- Navigate (
cd) to the main directory in pro_obo_slim - Run
sh odk.sh make allto download docker image and run code to update the Slim version of PR
At this point I had to stop, because the diff has added ^^xsd:string statements. This is a known issue:
- https://obophenotype.github.io/cell-ontology/Fixing_xsdstring_diffs/
But I am not sure on how to do it for pro_obo_slim
Hi @lubianat, do you use macos? You can try
sed -i '' -E "s/Annotation[(](oboInOwl[:]hasDbXref [\"][^\"]*[\"])[)]/Annotation(\1^^xsd:string)/g" pr_slim.owl
@lubianat, if you merge this PR by end of day today, it can be incorporated for the next CL release, which is planned for tomorrow. Otherwise, it would go in a future release.
This PR has not seen any activity in the past month; if nobody comments or reviews it in the next week, the PR editor will be allowed to proceed with merging without explicit approval, should they wish to do so.
Hi, @lubianat, can we close this PR? If there are pending tech issues, can you work with @anitacaron to resolve?
@bvarner-ebi I am sorry for this delay. All the technical hiccups got me a bit demotivated with contributing to CL.
I will try again to update pro_obo_slim, but I am finding it very hard to contribute on a regular-but-not-time-consuming basis. Some comments below on clarifying my position:
This issue has already taken more than 2 years (!), while on Wikidata I could add it as soon as I desired, taking in a couple minutes on June 2021.
This task, for example, requires a lot of memory usage; I am lucky to have 32 GB RAM in my work computer, 16 wouldn't cut it. It would be good if the hardware requirements were documented; I risked a crash in my PC and losing other stuff I am running.
Thank you @anitacaron for the help, but for some reason the MacOS macro did not work in Ubuntu (though sed is supported). I frankly feel a bit lost. Should I open an issue in PRO OBO SLIM, add a PR and try and fix things there?
CL is a great effort and tool, and I greatly appreciate the community.
However, given my technical skill set, and the current gaps in documentation, I will restrict my contributions to very simple edits so as to keep motivated, and keep helping, at least for the while.
As so, I won't touch more this PR and this issue, and on my side, you may close it as undone.
@lubianat, thank you for the update.
May I suggest going forward to use the new term request template and leaving it unassigned?
That way you can still contribute to CL by providing content in the template and an editor can pick up the issue and manage the technical aspects of adding the term. It's an option to continue contributing to CL without navigating the potential technical hurdles.
[EDITED], @anitacaron or @shawntanzk to assist.
Hi Tiago,
This issue has already taken more than 2 years (!), while on Wikidata I could add it as soon as I desired, taking in a couple minutes on June 2021.
This task, for example, requires a lot of memory usage; I am lucky to have 32 GB RAM in my work computer, 16 wouldn't cut it. It would be good if the hardware requirements were documented; I risked a crash in my PC and losing other stuff I am running.
I agree that this is completely ridiculous. I also think it is unnecessary. You should be able to do your edits and hand this off to the technical team to do relevant updated before a release.
There is a very simple fix for referring to terms not yet in imports - that is just add a new Class in Protege under root, pasting in the URL (https://oboacademy.github.io/obook/howto/update-import/#protege-based-declaration). This has been discouraged in recent years because it can cause some pipeline issues, but that should not be something anyone outside of the technical team should need to worry about.
Hope this helps, David
@anitacaron to fix the PRO references on this ticket so that it can be merged.
I'm a bit worried about making a equiv axiom as i don't know how strong the markers are, so I just did subclasses instead, hope that is ok. @lubianat if you are alright with this, please go ahead and merge. Thanks