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] new Bio API

Ray Fergerson ray.fergerson at
Wed Feb 26 14:17:33 PST 2014



Thanks. This is very helpful. The metrics are available for any particular
ontology with the call:


This call is always very fast because the metrics are all precalculated.
If you have a fixed set of ontologies this this is what you should use.


If instead you want to loop over all ontologies and look at their metrics
(this sounds like what you want) then you need to make a single call to
get back all ontology objects and then traverse to the "metrics" link in
each and make that call for each ontology (the URL there is the same form
as the one above but there is no reason to build it yourself since we have
already done so.


and then traverse to the "metrics" link on each ontology.


It is easiest to look at this in a browser like Chrome and see the results
and then generate the code accordingly.


This should all work fairly quickly. The list of ontologies may take a few
seconds to retrieve but the metrics calls are all sub-second.


We would be happy to hear more success stories about the annotator or
BioPortal in general. If you can point us to a talk or journal article
that we can read that would be great!





From: Bob Stanley [mailto:rstanley at] 
Sent: Wednesday, February 26, 2014 12:21 PM
To: Ray Fergerson; Sergey Nikitin; support at
Cc: musen at; Erich Gombocz
Subject: RE: new Bio API


Hi Ray,


Sergey is copying me on these notes so I thought I'd chime in from the
user side.  One use case is to run a relatively low-bandwidth query from
within our system to return metrics on a specific ontology of interest
before attempting to actually download or import it. With a targeted query
that completes quickly and brings back ontology metrics (# of classes,
etc.), we can get a (admittedly limited, but) better idea for what may
actually be there. This allows to avoid putting unnecessary burdens on the
environment that can be caused by downloading what may turn out to
extremely large, incomplete or un-useful ontologies. Open to other
concepts to make it easier to browse useful metrics on selected ontologies
or the entire list.


Although we've chatted internally with Sergey about some of the above, I
wanted to write you directly also to note that we're also recently
exercising Annotator more actively with some positive results.


I'll be mentioning these contributions by Stanford (BioPortal, Annotator)
in my talk at the Boston conference on semantic technologies in healthcare
/ life sciences (CSHALS) this week. 


Thanks for all your help!


All the best,





From: Ray Fergerson [mailto:ray.fergerson at] 
Sent: Tuesday, February 25, 2014 2:57 PM
To: Sergey Nikitin; support at
Cc: Bob Stanley; musen at; Erich Gombocz
Subject: RE: new Bio API




This information doesn't get expanded anymore because we thought that very
few users would actually want it and it greatly slows down the call for
everyone else.


Can you tell me what your use case is for needed the user information for
all ontologies? This doesn't seem very common. As an aside, since the user
data only very rarely changes, you could instead just get the all user
information and cache it on your side. You wouldn't need to retrieve it on
every call.




From: Sergey Nikitin [mailto:snikitin at] 
Sent: Tuesday, February 25, 2014 2:20 AM
To: support at
Cc: Bob Stanley; musen at; Erich Gombocz;
ray.fergerson at
Subject: new Bio API


Hi Support, 


    In the old API the ontologies request returned a lot of information
about each item. We are trying to find a way to receive alomost the same
results with the new API. 


            In the new API one can receive an ontology list in json format
using URL:{api-key}


    The result structures contain links to other objects, like:


"administeredBy": [





     Is there a way to receive ontologies with data obtained by following
the links in one request? For example:


"administeredBy": [ <> "

.         username": "pjd",

.         "email": "pdevries at",

.         "role": [ 



.         "@id":  <>

.         "@type":  <>


"@context": { 

o    "@vocab":  <>




 Perhaps, this is not the right approach? Then, what you recommend? loads pretty fast. We want
to achive the same result.



Sergey Nikitin

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

More information about the bioontology-support mailing list