Gene matching and entry link?

[buske]

We've had requests from users for gene matching functionality from the Monarch MME endpoint. They've said this is the most important functionality for them to assess model organism and known disease data. They've also requested having the contact URL being a link to more information on Monarch about that entry. Are either of these possible?

[jmcmurry]

@cmungall is the best positioned to reply here but he is traveling. In the meantime, perhaps I can help. "Links to more information in Monarch" can be manually derived if you know the prefixed accession of the concept; you can omit type and it will resolve for you. For instance https://monarchinitiative.org/NCBIGene:6622 resolves to https://monarchinitiative.org/gene/NCBIGene:6622.

Moreover, you don't even have to know which of the identifiers is the 'authoritative' one. As long as we have them formally represented as equivalent identifiers, those too will resolve for you. For instance: https://monarchinitiative.org/UMLS:C0022593 resolves to https://monarchinitiative.org/disease/DOID:161.

A list of all the prefixes we use can be found here.

Does this help in the meantime?

[cmungall]

I think it's more about this part of the API
https://github.com/ga4gh/mme-apis/blob/v1.0a/search-api.md#genomicfeatures

On 31 Aug 2016, at 19:16, Julie McMurry wrote:

@cmungall is the best positioned to reply here but he is traveling. In
the meantime, perhaps I can help. "Links to more information in
Monarch" can be manually derived if you know the prefixed accession of
the concept; you can omit type and it will resolve for you. For
instance
https://monarchinitiative.org/NCBIGene:6622
resolves to
https://monarchinitiative.org/gene/NCBIGene:6622.

Moreover, you don't even have to know which of the identifiers is the
'authoritative' one. As long as we have them formally represented as
equivalent identifiers, those too will resolve for you. For instance:
https://monarchinitiative.org/UMLS:C0022593 resolves to
https://monarchinitiative.org/disease/DOID:161.

A list of all the prefixes we use can be found
here.

Does this help in the meantime?

You are receiving this because you were mentioned.
Reply to this email directly or view it on GitHub:
#18 (comment)

[buske]

Sorry for the slow follow-up. Regarding the follow-up link, yes, that's helpful to know @jmcmurry. We can certainly hard-code this into the Monarch user interface for PhenomeCentral. Does it make more sense for this URL to be included as the contact URL though, instead of the generic email (the href field of the contact object)?

The main concern though is still that gene information (the genomicFeatures data, as @cmungall mentioned) isn't used for matching.

[jmcmurry]

Does it make more sense for this URL to be included as the contact URL though, instead of the generic email (the href field of the contact object)?

Possibly, since from any given data page the user can reach the contact email anyway. Having said this we would not want to dissuade anyone from contacting us by email. We would welcome it, in fact. We can always point someone in the right direction from there.

[jnguyenx]

@buske We had some discussion here and the plan is to wait for the next version of one of our tool, owlsim3. With the next version it'll be easier for us to provide gene matching.

Unfortunately I don't have any time estimate to provide at this point.

[buske]

Thanks for the update. Sounds great!