RE: URIs for UMLS

With regard to URIs for vocabulary entities, I have already completed this
task for the NLM MeSH 2010 thesaurus (see attached abstract for AMIA CRI
2010), and will be adding additional major vocabularies including UMLS in
the coming months.

For examples of URIs/URLs for MeSH2010 consider the following examples from
the PDS MeSH2010 web service:

http://pds.portaldoors.net/mesh2010/d000001

for a single Descriptor Record embedded in PDS Message wrapper, or

http://pds.portaldoors.net/mesh2010/resrep/search?nam=informatics

for a search that returns a set of records each with its own URI. A more
detailed paper on the PDS MeSH 2010 web service is in preparation.

Note that the general PDS framework allows for flexibility in distinguishing
URIs from URLs so that any entity identified by a URI may have multiple
locations that may include mirrored sites and/or different kinds of access.
However, in the case of the current PDS implementation of the MeSH2010
vocabulary, the convention is adopted that the URL is the same as the URI so
that it can be accessed readily from the web service which does not yet have
any mirrors.

Also, please note that I do include the statement as required by NLM about
NLM being the creator and maintainer of the original source Descriptor
Records. This statement appears in every response returned by the PDS
MeSH2010 web service.

Should anybody wish to discuss, I hope that you will visit with me in person
at AMIA CRI 2010 in San Francisco next month.

Regards,

Carl


***********************
Carl Taswell, MD, PhD
Global TeleGenetics, Inc.
8 Gilly Flower Street
Ladera Ranch, CA 92694
Tel: 949-481-3121




-----Original Message-----
From: public-semweb-lifesci-request@w3.org
[mailto:public-semweb-lifesci-request@w3.org] On Behalf Of Kei Cheung
Sent: Thursday, February 11, 2010 8:00 PM
To: Matthias Samwald
Cc: public-semweb-lifesci
Subject: Re: URIs for UMLS

May be a related question, for gene information, should be use entrez 
gene id or umls id (cui)?

Cheers,

-Kei

Matthias Samwald wrote:

> Sorry for asking such a seemingly simple question. Establishing URIs 
> for UMLS entities has now been discussed for years. What is the 
> current status of this development? Do we have somehow useful, 
> acccepted, possibly linked-data friendly entities for UMLS IDs by now? 
> I seem to be unable to find any documentation of such a thing, only 
> proposals in research papers.
>  
> Cheers,
> Matthias