New link outs to GeneDB

[pgaudet]

Hi,

GeneDB now links to the EuPathDB portal. I made a PR here to update the link out info:
#1778
but I probably did something not allowed for the yaml file.

The data should link to one of two URLs depending on the species:
Plasmodium:
https://plasmodb.org/plasmo/app/record/gene/PF3D7_1467300

L. major and T. brucei
https://tritrypdb.org/tritrypdb/app/record/gene/Tb927.1.630
https://tritrypdb.org/tritrypdb/app/record/gene/LmjF.12.1310

[kltm]

@pgaudet Is this not included in the work being described in #1778 and #1777 ?

[pgaudet]

#1778 is the PR for this ticket here (link outs to EuPathDB).

#1777 is about data sources.

[pgaudet]

Hi @kltm

I was trying to keep the data sources and the link outs (db-xrefs.yaml) separate.

For link outs, you wrote here #1777 (comment)
you wrote

From #1778, this inconsistent:

      url_syntax: http://www.genedb.org/gene/[example_id]
      example_id: PF3D7_1467300
      example_url: http://www.genedb.org/gene/PF3D7_1467300
      example_url: https://plasmodb.org/plasmo/app/record/PF3D7_1467300
      example_id: Tb927.1.630
      example_url: https://tritrypdb.org/tritrypdb/app/record/gene/Tb927.1.630
      example_id: LmjF.12.1310
      example_url: https://tritrypdb.org/tritrypdb/app/record/gene/LmjF.12.1310     

If the url_syntax is http://www.genedb.org/gene/[example_id], only http://www.genedb.org/gene/PF3D7_1467300 can be valid. If there are different URLs/structures, they need to be listed as different databases.

---

Is it OK if we create two entries, one for PlasmoDB and one for TriTrypDB?

database: PlasmoDB
name: PlasmoDB
rdf_uri_prefix: https://identifiers.org/plasmodb
generic_urls:
- https://plasmodb.org
entity_types:
- type_name: gene
type_id: SO:0000704
id_syntax: PF3D7_[0-9]{7})
url_syntax: https://plasmodb.org/plasmo/app/record/[example_id]
example_id: PF3D7_1467300
example_url: https://plasmodb.org/plasmo/app/record/PF3D7_1467300

database: TriTrypDB
name: TriTrypDB
rdf_uri_prefix: https://identifiers.org/tritrypdb
generic_urls:
- https://tritrypdb.org/
entity_types:
- type_name: gene
type_id: SO:0000704
id_syntax: ((LmjF|LinJ|LmxM).[0-9]{2}.[0-9]{4})|(Tb[0-9]+.[A-Za-z0-9]+.[0-9]+)|(Tb.[0-9]{6}.[0-9]+)
url_syntax: https://tritrypdb.org/tritrypdb/app/record/gene/[example_id]
example_id: Tb927.1.630
example_url: https://tritrypdb.org/tritrypdb/app/record/gene/Tb927.1.630
example_id: LmjF.12.1310
example_url: https://tritrypdb.org/tritrypdb/app/record/gene/LmjF.12.1310

I'll make a new PR with this.

Thanks, Pascale

[Achchuthan]

Hi Pascale,

database: PlasmoDB name: PlasmoDB rdf_uri_prefix: https://identifiers.org/plasmodb generic_urls: - https://plasmodb.org entity_types: - type_name: gene type_id: SO:0000704 id_syntax: PF3D7_[0-9]{7}) url_syntax: https://plasmodb.org/plasmo/app/record/[example_id] example_id: PF3D7_1467300 example_url: https://plasmodb.org/plasmo/app/record/PF3D7_1467300

url_syntax should be [https://plasmodb.org/plasmo/app/record/gene/[example_id]]
example_url should be https://plasmodb.org/plasmo/app/record/gene/PF3D7_1467300

database: TriTrypDB name: TriTrypDB rdf_uri_prefix: https://identifiers.org/tritrypdb generic_urls: - https://tritrypdb.org/ entity_types: - type_name: gene type_id: SO:0000704 id_syntax: ((LmjF|LinJ|LmxM).[0-9]{2}.[0-9]{4})|(Tb[0-9]+.[A-Za-z0-9]+.[0-9]+)|(Tb.[0-9]{6}.[0-9]+) url_syntax: https://tritrypdb.org/tritrypdb/app/record/gene/[example_id] example_id: Tb927.1.630 example_url: https://tritrypdb.org/tritrypdb/app/record/gene/Tb927.1.630 example_id: LmjF.12.1310 example_url: https://tritrypdb.org/tritrypdb/app/record/gene/LmjF.12.1310

We have updated Leishmania infantum (LinJ) genome with a newer assembly in TriTrypDB and the identifiers are now different. An example is LINF_330027200. So, the id_syntax for this genome should be [LINF_[0-9]{9}].

Thanks,
Achchuthan

[pgaudet]

Thanks, I made the changes, let me know if that doesnt seem right.

[pgaudet]

• Add GeneDB back to dbxref while PlasmoDB and TriTryDB are being sorted (need to fix the namespaces)
• Remove ApiDB_PlasmoDB

[pgaudet]

For linking out- exchanged with @Achchuthan on Feb 2

Hi Ach,

One other request (hopefully we’re done after this): for the links to resolve correctly, there are 2 options:

1. You change the GAF files to have the correct prefix in Column 1: PlasmoDB/TriTryDB, rather than GeneDB.
2. You keep GeneDB in column 1, but then you need some redirect system on the EuPathDB website.

Do you think one of these 2 options can be done?

Thanks, Pascale

---

@Achchuthan 's response:
Hi Pascale,

I don’t think option 1 is possible as far as the Sanger FTP files are concerned. If we are to do the change with GAF files, we need to change the location to somewhere else. Sanger FTP files are automatically generated and we don’t have any developer support from Sanger anymore.

I don’t quite understand option 2. But, we reference GeneDB as the source in PlasmoDB/ TriTrypDB.

Best wishes,
Achchuthan

[pgaudet]

This is not yet resolved (link to GeneDB cannot be automatically built from the information in the genedb yaml file).

Ach has left the project - Ulrike Boehme and Beatrice Amos are now taking care of the project.

[kltm]

Closed for #2256