consider xrefs to MedlinePlus and GeneReviews

[nicolevasilevsky]

@galyea123 emailed about this question:

I thought to ask you if it will be possible to add other links (ref) to your has dbxref section. Specifically, MedlinePlus which has patient-friendly information and GeneReviews, which have detailed information. Note that MedlinePlus and GeneReviews are already tagged in OMIM. Is this something you are able to do?

cc @sabrinatoro @matentzn

[galyea123]

Thank you Nicole!

[kanems]

FYI MedGen also has these links and uses GeneReviews for our primary definitions whenever possible. See example - https://www.ncbi.nlm.nih.gov/medgen/?term=C0410179

[matentzn]

@nicolevasilevsky @sabrinatoro this request is deeply tied to the contentious #7071 issue - that needs to be solved first, and I think what it really needs is a brave decision. Its one of these things where 49% of the team will say a relatively hard "no" due to maintenance overhead and lack of separation of concerns ("an ontology is not a linkout database"), and 51% of the team will say "why not? Its valuable to users and easy to add - let's add it!". Someone needs to make a call!

[galyea123]

[like] Gioconda Alyea reacted to your message:

…

________________________________ From: Nico Matentzoglu ***@***.***> Sent: Monday, March 25, 2024 11:12:46 AM To: monarch-initiative/mondo ***@***.***> Cc: Gioconda Alyea ***@***.***>; Mention ***@***.***> Subject: Re: [monarch-initiative/mondo] consider xrefs to MedlinePlus and GeneReviews (Issue #7425) @nicolevasilevsky<https://github.com/nicolevasilevsky> @sabrinatoro<https://github.com/sabrinatoro> this request is deeply tied to the contentious #7071<#7071> issue - that needs to be solved first, and I think what it really needs is a brave decision. Its one of these things where 49% of the team will say a relatively hard "no" due to maintenance overhead and lack of separation of concerns ("an ontology is not a linkout database"), and 51% of the team will say "why not? Its valuable to users and easy to add - let's add it!". Someone needs to make a call! — Reply to this email directly, view it on GitHub<#7425 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/A4QAWD4BOOZ4UDSFOEQTIJTY2ABC5AVCNFSM6AAAAABE4HJR3GVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAMJXG43DCMZYGQ>. You are receiving this because you were mentioned.Message ID: ***@***.***> Any review, use, distribution or disclosure by others is strictly prohibited.

[galyea123]

Hi Sabrina! Are there any news about this request?

Note: We are planning on having relevant sources appearing in every RD when we launch the NORD RD list live.

Have a great holyday! Thank you, Gioconda Alyea

[sabrinatoro]

Hi @galyea123,
We have discussed this and need to talk about implementation. I think we were hoping to get the cross-links from MedGen (since they already have them). I will talk with @matentzn to see where we are in the process.
Thank you! Have a nice holiday too!

[galyea123]

Thank you Sabrina!

…

________________________________ From: Sabrina Toro ***@***.***> Sent: Wednesday, July 3, 2024 1:59 PM To: monarch-initiative/mondo ***@***.***> Cc: Gioconda Alyea ***@***.***>; Mention ***@***.***> Subject: Re: [monarch-initiative/mondo] consider xrefs to MedlinePlus and GeneReviews (Issue #7425) Hi @galyea123<https://github.com/galyea123>, We have discussed this and need to talk about implementation. I think we were hoping to get the cross-links from MedGen (since they already have them). I will talk with @matentzn<https://github.com/matentzn> to see where we are in the process. Thank you! Have a nice holiday too! — Reply to this email directly, view it on GitHub<#7425 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/A4QAWD3N5VHBDJAWW2TEJJ3ZKQ3YDAVCNFSM6AAAAABE4HJR3GVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDEMBWHEYDOMZWHE>. You are receiving this because you were mentioned.Message ID: ***@***.***> Any review, use, distribution or disclosure by others is strictly prohibited.

[matentzn]

Depends on

• Add genereview links from MedGen FTP to release output has ROBOT template medgen#29 (@twhetzel)

[kanems]

FYI MedlinePlus has a ton of developer tools that may be easier than trying to get their data via MedGen (which we process but haven't put out in FTP reports at this time) https://medlineplus.gov/about/developers/

[matentzn]

@kanems thanks a ton.

@twhetzel @galyea123 in this case (monarch-initiative/medgen#29 (comment)), I am not right now convinced this should be made a priority. Can we create a strong case for including these linkouts in to Mondo? Does anyone in the community if the resource to prepare a table for us to import into Mondo? (Mondo ID, xref)

[kanems]

In following up with this issue and the ping about it again during the Mondo call, I was able to pull a one-off report together for you all based on a couple of FTP reports from NCBI and my very ROUGH working memory of R to try and clean up a report based on the matching based on IDs...
https://ftp.ncbi.nlm.nih.gov/pub/GeneReviews/ GeneReviews FTP site includes a report of all their NBK IDs and chapter shortnames based on OMIM mappings.
Obviously Mondo team is familiar with the MedGen ID mappings file that also includes OMIM #s and MedGen UIDs, CUIs, etc.

So I pulled the GeneReviews report NBKid_shortname_OMIM.txt and then used matching across the various tables on CUI and MIM numbers to get the MIM number, CUI, MedGen UID, NBK ID, disease name, GeneReviews Short name and the formatted URL to the specific genereviews chapter on NCBI bookshelf.
There are 1992 items here and NO mondo IDs (sorry, I did not think to have to merge in the Mondo IDs as well). The Full GR FTP report has more MIM#s if you all want to try and increase your coverage with that file (NBK-MIM-Mondo)
MedGenIDs_toGeneReviews.xlsx

[sabrinatoro]

Thank you very much @kanems !!!
I am tagging the technical team (@twhetzel and @matentzn though he is out at the moment), so they can take it from here.

[matentzn]

Fantastic @kanems! While it would, thanks to your work, be super easy to pull the gen-reviews links in as a one-off, I am not sure this is a good idea without a maintenance plan; Could you see yourself adding the table to your MedGen release system to keep it up to date, or is that out of scope for you?

[kanems]

@matentzn I will discuss with our devs and see what we can do. It would probably be a few months before we'd release an updated report (if we incorporate it at all, I am not the one to make that call).
The GeneReviews FTP report linked above plus the existing ID mapping report from MedGen is all you need to match a gene reviews NBK ID to a CUI via MIM number. Since Mondo has MIM #s already, you could probably match the GeneReviews FTP report by MIM# directly.

[galyea123]

The same is true for MedlinePlus. OMIM has the #s already so you could match these from OMIM. I will also talk with Sean as we are doing this in NORD. Thank you! Gioconda