Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Provide mapping between EFO response to terms and new OBA replacements #1436

Open
dosumis opened this issue Feb 2, 2022 · 12 comments
Open
Assignees

Comments

@dosumis
Copy link
Collaborator

dosumis commented Feb 2, 2022

Generate list of all GO classes in the response to branch - this can be done easily with a query of UberGraph.

SPARQL query for xrefs to these terms => EFO terms.

Draft query here - needs some work.

TODO assign to @rays22 (needs to be added to EFO)

@rays22
Copy link
Collaborator

rays22 commented Feb 3, 2022

Here is an UberGraph query
to get the EFO terms that are subclasses of

  • GO:0042221 ; # "response to chemical"
  • GO:0050896 ; # "response to stimulus"
    Feel free to modify it.

@dosumis dosumis added this to the GWAS April/May milestone Apr 29, 2022
@dosumis
Copy link
Collaborator Author

dosumis commented Apr 29, 2022

  1. We need a mapping between OBA terms and EFO/GO terms in use so that we can obsolete and replace.  For this, we can use Kalia’s mapping spreadsheet - but note, the OBA IDs in this or NOT correct.  Instead we can use CHEBI IDs. e.g. 


defined_class defined_class_name chemical chemical_name  
OBA:2040000   CHEBI:10033 warfarin  

As a first step, we should create a 3 column spreadsheet:  OBA to EFO to GO mapping.


We need a mapping between OBA terms and EFO/GO terms in use so that we can obsolete and replace. For this, we can use [Kalia’s mapping spreadsheet](https://docs.google.com/spreadsheets/d/13qh7dLE38vMyz91oRqj6GzKjFohNazNKAJxKG6Plw1o/edit#gid=0) - but note, the OBA IDs in this are NOT correct. Instead we can use CHEBI IDs. e.g.
defined_class defined_class_name chemical chemical_name
OBA:2040000 CHEBI:10033 warfarin

Kallia’s spreadsheet has the mapping:

EFO:0009167 response to warfarin GO:0042221 CHEBI:10033

From this we can map OBA:2040000 to
EFO:0009167 and GO:0042221

As a first step, we should create a 3 column spreadsheet: OBA to EFO to GO mapping.

@rays22
Copy link
Collaborator

rays22 commented May 4, 2022

@dosumis
Copy link
Collaborator Author

dosumis commented May 4, 2022

DEPENDENCY: OBA release needed.

  • Ray to add an additional column showing obsoletion status to EFO/GO--> OBA mapping for response to chemical terms.

@zoependlington everything with SKOS:exactMatch in this sheet is good, but I think more discussion of transition is needed to ensure that labels match up and the imported terms sit in the right place in the hierarchy.

@rays22
Copy link
Collaborator

rays22 commented May 24, 2022

  • Ray to add an additional column showing obsoletion status to EFO/GO--> OBA mapping for response to chemical terms.

I have replaced the obsolete EFO classes with the replacement GO classes in the gsheet tab.
Now there should be no obsolete EFO/GO classes in the table.

@rays22
Copy link
Collaborator

rays22 commented Jun 8, 2022

@dosumis , @zoependlington ,
Is there anything else to do to for this ticket or can I close it?

@dosumis
Copy link
Collaborator Author

dosumis commented Jul 18, 2022

The aims here has to be to get OBA response_to terms seamlessly integrated into EFO. For this I think we need a mapping sheet for role-based grouping classes as well as the ones here gsheet tab.

I think what's needed it:

  • import all OBA response_to terms to EFO
  • Using this mapping gsheet tab. + one for role-based grouping classes:
    • where we have an exact match between an OBA term and an EFO term, the EFO term should be obsoleted in favour of the OBA term. @zoependlington - I think we've agreed this but please let me know if problematic
    • where the EFO term is more specific (broad_match in the table)? it should be retained as subclass
    • where the EFO term is broader (narrow_match in the table?) it should be retained as a superclass.

This would be rather painful to do by hand. @matentzn - I'm wondering if there's a clever way to do this. It would probably take a few attempts to the tables right and review the results.

@matentzn
Copy link
Contributor

Too complicated to answer in a ticket, needs meeting

@rays22
Copy link
Collaborator

rays22 commented Aug 8, 2022

2022-08-08:

  • raise this issue at the next OBA call.

@dosumis
Copy link
Collaborator Author

dosumis commented Oct 12, 2022

I think it will take a call to fix it - with some prep so that we can see plenty of examples of what. We have the mapping table - but this is a bit confusing to look at on its own. It would be good to have some examples where we can see the representation in OBA and EFO side by side.

Also before the call - @rays22 - can we get a status-update on the role-based terms? Are they now in OBA? Is there more work to do on them?

I propose that we discuss this on the next OBA call and invite Zoe along.

@rays22
Copy link
Collaborator

rays22 commented Oct 12, 2022

Also before the call - @rays22 - can we get a status-update on the role-based terms? Are they now in OBA? Is there more work to do on them?

Some role-based terms related toresponse_to_chemical_with_role_stimulus have been incorporated into OBA.

@dosumis
Copy link
Collaborator Author

dosumis commented Mar 13, 2023

I can't fix this simply with a response on the ticket. We need a call to discuss a general strategy for incorporating OBA terms into EFO where there are existing EFO terms to integrate with that are not in OBA.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

4 participants