Search Mailing List Archives

Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[bioontology-support] ATC / bioportal

Jennifer Leigh Vendetti vendetti at
Tue Jan 10 17:56:16 PST 2017

I’m not sure.  That question would be best posed to the folks at the National Library of Medicine (UMLS support page:

On Jan 10, 2017, at 12:23 PM, Alexander Garcia Castro <alexgarciac at<mailto:alexgarciac at>> wrote:

could the difference be due to the transformation from the ATC original format to the  UMLS format?

On Tue, Jan 10, 2017 at 3:13 PM, Jennifer Leigh Vendetti <vendetti at<mailto:vendetti at>> wrote:
Hello Olga,

On Jan 9, 2017, at 8:38 PM, Olga Ximena Giraldo <oxgiraldo at<mailto:oxgiraldo at>> wrote:

If I understand correctly you do not take care of converting (transforming) the original WHO ATC vocabulary into an RDF file. Instead, you just take whatever UMLS has and that is what you transform to RDF. Right?

Yes, that’s correct.

The original WHO ATC, as available at [1]  does not have the classes “Ergot alkaloid oxytocics” and “Porstaglandins, Oxytocics”
while the one in bioportal does have these two classes [2], [3].

You are seeing those classes because they are present in the UMLS distribution of ATC that we imported into BioPortal (UMLS version 2016AA).  I’ve included a couple of screenshots below from the UMLS Metathesaurus Browser where you can execute searches restricted by source ontology and release version.

I would like to know what other differences in the hierarchy are there between the original WHO ATC (14 main groups) and that in bioportal (17 main groups) [4].

The version of ATC contained in the UMLS has 17 root classes - that’s why you see the same in BioPortal (another screenshot below of the ATC hierarchy from the UMLS Browser).  We don’t modify the contents of the ontology when we do the import.

I’m assuming the website you mention below ( offers the most current version of ATC (although I didn’t see a version number anywhere)?  If you’re seeing differences between the canonical distribution from the WHO and BioPortal content, it’s likely because BioPortal has an older version.  According to UMLS documentation, ATC is only updated in their system once annually:

Kind regards,

thanks in advance,


<Screenshot 2017-01-10 11.33.35.png>

<Screenshot 2017-01-10 11.35.02.png>

<Screenshot 2017-01-10 11.43.42.png>

bioontology-support mailing list
bioontology-support at<mailto:bioontology-support at>

Alexander Garcia

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <>

More information about the bioontology-support mailing list