Details
-
New Feature Request
-
Resolution: Canceled
-
Major
-
None
-
MONARCH 1.0
-
None
-
NIF
-
Issues closed as MONARCH has transitioned from UCSD services
Description
ClinVar uses their own vocabulary of traits, which they have internal identifiers for. They do make mappings to MedGen and other vocabularies too. Note however that sometimes the MedGen mappings are missing (usually a synchrony issue), and eventually come later; I believe MedGen identifiers are their official external id.
So, do we want to add the "internal" identifiers for ClinVar traits into our disease ontology (when they are diseases)?
Here is the view over that data. http://beta.neuinfo.org/mynif/search.php?q=*&t=indexable&nif=nlx_151671-4
You can't see the trait ids, but they are in the underlying data...see for example:
http://beta.neuinfo.org/services/v1/federation/data/nlx_151671-4.json?exportType=data
Alternatively, we should use their MedGen to OMIM mappings. Note that MedGen is mostly just UMLS ids (if starting with C), but some are unique to MedGen (prefixed with CN).
Also note that the medgen-to-omim (and other vocab) mappings here can't be trusted as equivalence classes. For example, for Hermasky-Pudluck syndrome,
http://beta.neuinfo.org/mynif/search.php?q=DOID_3753&dis=Hermansky-Pudlak%20syndrome&t=indexable&nif=nlx_151671-4&b=0&r=20
it has several subtypes; each one maps to GeneReviews:NBK1287, which is for a general type of Hermansky-Pudlak.