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] UMLS semantic types in NCBO Annotator and SPARQL endpoint

Jael GarcíaCastro leyla.jael.garcia at gmail.com
Tue Oct 8 13:25:01 PDT 2013


Thanks Paul,

Could you please point me to the documentation about the API
data.bioontology.org? In rest.bioontology.org, the parameter for the text
to be annotated used to be "textToAnnotate" but I saw it is now "text" in
data.bioontology.org. I would like to check the rest of parameters to make
sure I am using it in the proper way.

Thanks again,
Jael


On Tue, Oct 8, 2013 at 9:11 PM, Paul R Alexander <palexander at stanford.edu>wrote:

> Jael,
>
> The new API code is released at data.bioontology.org, so you should point
> there and not stagedata.bioontology.org (this is our testing platform).
>
> rest.bioontology.org is essentially deprecated as of now, so new fixes
> won't end up there.
>
> There is indeed a bug on the xml output at data.bioontology.org that is
> causing the class information to not appear. This will be fixed later
> today. If you need to see the output now, you can try the json
> serialization. The xml will look different than the xml provided at
> rest.bioontology.org, primarily because of the transition to our new API,
> which is now more uniform across individual NCBO APIs.
>
> Please let me know if you have further questions.
>
> Paul
>
>
> On Oct 8, 2013, at 1:04 PM, Jael GarcíaCastro <leyla.jael.garcia at gmail.com>
> wrote:
>
> Dear Ryan,
>
> I have managed to work with the URL
> http://rest.bioontology.org/obs/annotator from Java, it is working fine.
> However, the problem with MeSH and the Semantic types persist here. I get
> T999 as semantic type when annotating with MeSH (virtual id 1351) while
> with SNOMED-CT (virtual id 1353) I do get T109 and T124. When will it
> solved?
>
> And the XML that I get in Java is different from the one that I get from
> the Annotator interface, the last one seems to be incomplete.
>
> Thanks,
> Jael
>
>
> On Tue, Oct 8, 2013 at 8:33 PM, Jael GarcíaCastro <
> leyla.jael.garcia at gmail.com> wrote:
>
>> Dear Ray,
>>
>> I have seen that Bioportal has been migrated now to the RDF platform. Now
>> "serotonin" in MeSH indeed has types T109 and T124 as you mentioned in your
>> previous mail, great!
>>
>> However, I am not getting the information I used to with NCBO Annotator.
>> Before the migration, I used this URL for the annotator
>> http://rest.bioontology.org/obs/annotator, now, I saw in the
>> documentation (
>> http://stagedata.bioontology.org/documentation#nav_annotator) that it
>> should be http://stagedata.bioontology.org/annotator, is that right? In
>> both cases I am getting an error saying that my API Key is not valid. I
>> already re-registered myself in case it has expired or kind of, but still I
>> get the same error with the new API Key. What am I doing wrong? Any
>> suggestions?
>>
>> Using the apikey used by the annotator interface when XML format is
>> required (
>> http://data.bioontology.org/annotator?text=serotonin&max_level=0&ontologies=MESH&format=xml&apikey=xxx),
>> with the URL http://rest.bioontology.org/obs/annotator I get a "Method
>> Not Allowed", while with http://stagedata.bioontology.org/annotator I
>> get an "Internal Server Error". What could be the problem?
>>
>> I tried to get the XML from the Annotator interface and the XML now is
>> different from before, it has less information, is this a new response?
>>
>> <annotationCollection><annotation><annotatedClass/><hierarchyCollection/><annotationsCollection><annotations><from>1</from><to>9</to><matchType>PREF</matchType><text>SEROTONIN</text></annotations></annotationsCollection><mappingsCollection/></annotation></annotationCollection>
>>
>>
>> Thanks so much in advance,
>> Jael
>>
>>
>> On Wed, Sep 25, 2013 at 3:19 AM, Ray Fergerson <
>> ray.fergerson at stanford.edu> wrote:
>>
>>> Jael,****
>>>
>>> ** **
>>>
>>> We may make the UMLS property URI’s that we generate resolvable but this
>>> is not a high priority at the moment.****
>>>
>>> ** **
>>>
>>> There does seem to be something wrong with the sematic types and MeSH.
>>> They work fine for other ontologies (e.g. SNOMED). By the way, Serotonin
>>> has types T109 and T124 (not T016 has stated below). These types are
>>> returned in the XML for “working” ontologies.****
>>>
>>> ** **
>>>
>>> Whatever the problem with MeSH, we are about to replace this entire
>>> backend system in a couple of days. Thus we are not going to look into this
>>> problem. The new system should work correctly. If it doesn’t then we will
>>> fix it. Stay tuned.****
>>>
>>> ** **
>>>
>>> Ray****
>>>
>>> ** **
>>>
>>> *From:* bioontology-support-bounces at lists.stanford.edu [mailto:
>>> bioontology-support-bounces at lists.stanford.edu] *On Behalf Of *Jael
>>> GarcíaCastro
>>> *Sent:* Tuesday, September 24, 2013 12:09 PM
>>> *To:* support at bioontology.org
>>> *Subject:* [bioontology-support] UMLS semantic types in NCBO Annotator
>>> and SPARQL endpoint****
>>>
>>> ** **
>>>
>>> Hi all, ****
>>>
>>> I tried DESCRIBE <http://purl.bioontology.org/ontology/MDDB/07224> at
>>> http://sparql.bioontology.org. The description includes the UMLS
>>> semantic type as ****
>>>
>>> <http://bioportal.bioontology.org/ontologies/umls/hasSTY> <http://bioportal.bioontology.org/ontologies/umls/sty/T121>****
>>>
>>> ** **
>>>
>>> I noticed that the URL
>>> http://bioportal.bioontology.org/ontologies/umls/sty/T121 does not
>>> resolve. Do you plan these UMLS semantic type URLs to be resolvable?****
>>>
>>> I also tried the NCBO Annotator
>>> http://bioportal.bioontology.org/annotator with the term "serotonin"
>>> and the ontology MeSH. This term is associated to the UMLS semantic type
>>> T016. However, if I retrieve the annotation in the XML format using the
>>> option "Format Result As", the semantic type retrieved is T999 "NCBO
>>> BioPortal concept". I was expecting T016, is not the semantic type reported
>>> by the annotator the UMLS one?****
>>>
>>> Thanks,****
>>>
>>> Jael****
>>>
>>
>>
> _______________________________________________
> bioontology-support mailing list
> bioontology-support at lists.stanford.edu
> https://mailman.stanford.edu/mailman/listinfo/bioontology-support
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.stanford.edu/pipermail/bioontology-support/attachments/20131008/88e9fd4d/attachment-0001.html>


More information about the bioontology-support mailing list