Difference between revisions of "SNOMED CT OWL"
DavidWhitten (talk | contribs) (Created page with ""Jim McCusker" on Apr 16, 2019, 9:04 PM wrote: I looked at the SNOMED CT page: https://bioportal.bioontology.org/ontologies/SNOMEDCT expecting to find a link where you coul...") |
DavidWhitten (talk | contribs) |
||
(One intermediate revision by the same user not shown) | |||
Line 1: | Line 1: | ||
+ | "Jeffrey Staples" On Mon, Apr 15, 2019 at 4:35 PM wrote: | ||
+ | Hello [[Protégé]] users, | ||
+ | Could anyone point me an OWL version of SNOMED that I can load into Protégé? | ||
+ | |||
+ | ------ | ||
+ | "David Whitten" On Tue, Apr 16, 2019 at 3:53 PM wrote: | ||
+ | |||
+ | Jeffrey, | ||
+ | Assuming you mean SNOMED CT (and not SNOMED 2) when you find a copy I'd like one too. | ||
+ | |||
+ | It might stretch [[Protege]] a lot though. The wikipedia article claims over 311,000 concepts connected by over 1,360,000 links existed in 2011, so I'd like to see the OWL, but don't have much hope. Interestingly SNOMED CT is encoded in a subset of OWL called DL++ | ||
+ | |||
+ | ------ | ||
"Jim McCusker" on Apr 16, 2019, 9:04 PM wrote: | "Jim McCusker" on Apr 16, 2019, 9:04 PM wrote: | ||
− | I looked at the SNOMED CT page: https://bioportal.bioontology.org/ontologies/SNOMEDCT expecting to find a link where you could download the OWL file but couldn't find it. | + | I looked at the {{g|SNOMED CT}} page: https://bioportal.bioontology.org/ontologies/SNOMEDCT expecting to find a link where you could download the {{g|OWL}} file but couldn't find it. |
− | Michael DeBellis on Apr 16, 2019, 8:51 PM wrote: | + | "Michael DeBellis" on Apr 16, 2019, 8:51 PM wrote: |
I sent an email to their support and here is the response I just received: | I sent an email to their support and here is the response I just received: | ||
Line 27: | Line 40: | ||
[5] https://confluence.ihtsdotools.org/display/DOCTSG/9.2.6+SNOMED+CT+OWL+Distribution+FAQ | [5] https://confluence.ihtsdotools.org/display/DOCTSG/9.2.6+SNOMED+CT+OWL+Distribution+FAQ | ||
− | + | ------- | |
"Michael DeBellis" on Apr 16, 2019, 8:51 PM wrote: | "Michael DeBellis" on Apr 16, 2019, 8:51 PM wrote: | ||
Line 33: | Line 46: | ||
The short answer is that SNOMED-CT is NOT an open standard. This is a reason to decide whether to use it directly. SNOMED CT is not a subsumption hierarchy, it is a concept hierarchy. | The short answer is that SNOMED-CT is NOT an open standard. This is a reason to decide whether to use it directly. SNOMED CT is not a subsumption hierarchy, it is a concept hierarchy. | ||
− | + | ----- | |
"Jim Campbell" stated: | "Jim Campbell" stated: | ||
Latest revision as of 15:38, 17 April 2019
"Jeffrey Staples" On Mon, Apr 15, 2019 at 4:35 PM wrote: Hello Protégé users, Could anyone point me an OWL version of SNOMED that I can load into Protégé?
"David Whitten" On Tue, Apr 16, 2019 at 3:53 PM wrote:
Jeffrey, Assuming you mean SNOMED CT (and not SNOMED 2) when you find a copy I'd like one too.
It might stretch Protege a lot though. The wikipedia article claims over 311,000 concepts connected by over 1,360,000 links existed in 2011, so I'd like to see the OWL, but don't have much hope. Interestingly SNOMED CT is encoded in a subset of OWL called DL++
"Jim McCusker" on Apr 16, 2019, 9:04 PM wrote:
I looked at the SNOMED CT page: https://bioportal.bioontology.org/ontologies/SNOMEDCT expecting to find a link where you could download the OWL file but couldn't find it.
"Michael DeBellis" on Apr 16, 2019, 8:51 PM wrote:
I sent an email to their support and here is the response I just received:
The short answer is that the don’t have a SNOMED CT OWL file.
Jennifer wrote:
We import the SNOMED CT ontology into BioPortal from the UMLS Metathesaurus [1]. We developed Python scripts [2] that convert from a UMLS database to an RDF file that we can load into our application. Due to licensing restrictions, we aren’t permitted to allow downloads of the SNOMED CT RDF file, though the SNOMED CT data is fully available from the BioPortal application and our REST API [3].
Alternatively, you may be interested to look at the announcement that came out in Nov. of 2018 that SNOMED CT will be offered in OWL [4]. I poked around in their FAQ and it appears that the OWL file is something you have to generate yourself [5], as it’s not included in their core distribution.
Jennifer
[1] https://www.nlm.nih.gov/research/umls/knowledge_sources/metathesaurus/
[2] https://github.com/ncbo/umls2rdf
[3] http://data.bioontology.org/ontologies/SNOMEDCT
[5] https://confluence.ihtsdotools.org/display/DOCTSG/9.2.6+SNOMED+CT+OWL+Distribution+FAQ
"Michael DeBellis" on Apr 16, 2019, 8:51 PM wrote:
The short answer is that SNOMED-CT is NOT an open standard. This is a reason to decide whether to use it directly. SNOMED CT is not a subsumption hierarchy, it is a concept hierarchy.
"Jim Campbell" stated:
SNOMED CT is an international standard ontology for healthcare that is supported and maintained by 38 nations as members of the IHTSDO. It is a core terminology supporting interoperability in the Standards and Interoperability architecture of the ONCHIT. It is the largest and most comprehensive healthcare terminology/ontology worldwide. Such a resource is never free and every element of the US S&I architecture has been developed at a cost, generally from our tax dollars. SNOMED CT is open for growth and development through our own NLM and they maintain a terminology request service which manages and reviews requests for new content.
SNOMED CT has been released in an RF2 format for years but support for OWL has recently recently been proposed and is planned by the international community. In the US, any vendor or researcher can obtain SNOMED CT US edition free for use from the NLM with the simple mechanism of a UMLS license. The RF2 release from NLM can be converted to OWL format with a utility provided by the International community at:
https://github.com/IHTSDO/snomed-owl-toolkit
If any of our colleagues have trouble with that technically, we at Nebraska would be happy to provide them an OWL version once they verify they have obtained a UMLS license.