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] Search by code from REST API

John Graybeal jgraybeal at stanford.edu
Tue Feb 28 19:49:36 PST 2017


Hi Koray,

First, thanks for your observation about the ugly spam. We have had two reports, and don't have enough information to know whether this is someone who has joined the list, or is intercepting the content in some other way. We'll try to do some forensics to identify the problem and eliminate it.

Umm, taking your technical questions from the most sensitive and working backward:

Yes, we believe a whole lot of serious projects actually make use of BioPortal. We've been running a lot of analytics lately, and conservatively, we're seeing 35,000 page views weekly, and 11,000,000 API requests weekly (daily: 5000 pages/1.6 million API calls); these numbers are growing. We are confident these numbers are "real"; they exclude all known robots and anything with query behavior that looks like a robot, and there are a large number of unique users in both categories.

I'm going to go out on a limb here and say that most dev users who want to deal with specific (unique) concepts from ontologies will want to use the IRI that the ontology provider has specified as the unique identifier. (We consider the ontology itself the authority in this regard.)  This is how I would expect most researchers to "query by code"; it increases interoperability, and reduces the likelihood of missing content. (Because nothing prevents the same concept from being re-identified multiple times, with multiple 'unique identifiers', whether within identifiers.org<http://identifiers.org> or other services. Of course that works in the one direction because it resolves to the real identifier, but starts to fail if someone uses the identifiers.org<http://identifiers.org> IRI within some ontology.)

It appears that in the specific case of OPB, they have gone to the trouble of using identifiers.org<http://identifiers.org> to make the identifier in the ontology align with the identifier given to identifiers.org<http://identifiers.org> (different paths, but matching ID string at the end). However, BioPortal doesn't know what the full IRI is that identifiers.org<http://identifiers.org> is using; identifiers.org<http://identifiers.org> simply maps the unique (non-path) part of that identifier into a new path, namely the BioPortal purl path. And the term itself, in the ontology, has the actual unique identifier http://bhi.washington.edu/OPB#OPB_00340, which is the thing that is searchable in BioPortal.  So BioPortal doesn't know what prefix got assigned in Identifiers.org<http://Identifiers.org>; that capability is overlaid on top of the BioPortal system by an external system.

Would you be able to tell us more detail about your statement "pretty much all the biomedical research community is using identifiers.org<http://identifiers.org> IRIs"?  Am I correct that you are saying they are using the identifier.org<http://identifier.org> identifiers, instead of the identifiers specified in the ontologies? It would be great to understand the community that you are describing.

Thanks for relaying your use case, I hope this provides some helpful context.

John




On Feb 28, 2017, at 5:37 PM, Koray Atalag <k.atalag at auckland.ac.nz<mailto:k.atalag at auckland.ac.nz>> wrote:

Thanks for prompt reply,

An important issue: I’ve sent a previous support request without logging on to bioportal and immediately I started receiving SPAM email with explicit porn content! I’ve sent to the support email.

Back to my problem – I’m actually quite surprised that you don’t support this as pretty much all the biomedical research community is using identifiers.org<http://identifiers.org/> IRIs and that only points to the original organisation’s website usually; however the key to reach a particular term in a given ontology is given in the original URL. For example OLS from EBI supports this but they don’t have all the ontologies I need (e.g. OPB)

This does raise the question whether a lot of serious projects are actually making use of bioportal?? I think most dev users will want to query by code than just do text searching…Am I missing something here?

Cheers,

-koray

From: Michael Dorf [mailto:mdorf at stanford.edu]
Sent: Wednesday, 1 March 2017 2:22 p.m.
To: Koray Atalag
Cc: support at bioontology.org<mailto:support at bioontology.org>
Subject: Re: [bioontology-support] Search by code from REST API

Hi Koray,

Thank you for contacting us. I’m not sure whether I understood you entirely, but BioPortal allows searching by the full term id, and you can always limit the search by the ontology acronym (to eliminate instances of the same term in other ontologies:

http://bioportal.bioontology.org/search?q=http%3A%2F%2Fbhi.washington.edu%2FOPB%23OPB_00340&ontologies=OPB<http://bioportal.bioontology.org/search?q=http://bhi.washington.edu/OPB#OPB_00340&ontologies=OPB>

The equivalent API search is (you’ll need to add your apikey parameter):

http://data.bioontology.org/search?q=http%3A%2F%2Fbhi.washington.edu%2FOPB%23OPB_00340&ontologies=OPB<http://data.bioontology.org/search?q=http://bhi.washington.edu/OPB#OPB_00340&ontologies=OPB>

At the moment we don’t allow searches by the short IDs (i.e. OPB_00340).

Thanks!

Michael


On Feb 27, 2017, at 7:21 PM, Koray Atalag <k.atalag at auckland.ac.nz<mailto:k.atalag at auckland.ac.nz>> wrote:

Hi,

I’m trying to get prefLabel based on ontology identifier (e.g. in the form of identifier.org<http://identifier.org/> URIs)
I reckon you allow for searching by terms but not ontology codes – is there a way to do this?

Let’s say I have this URI: http://identifiers.org/opb/OPB_00340
So given I know I’m looking into OPB and code: 00340 I want to access the bioportal record and get prefLabel

For example using web interface I can get: http://bioportal.bioontology.org/ontologies/OPB?p=classes&conceptid=http%3A%2F%2Fbhi.washington.edu%2FOPB%23OPB_00340


Many thanks in advance

Cheers,

-koray

_______________________________________________
bioontology-support mailing list
bioontology-support at lists.stanford.edu<mailto:bioontology-support at lists.stanford.edu>
https://mailman.stanford.edu/mailman/listinfo/bioontology-support

_______________________________________________
bioontology-support mailing list
bioontology-support at lists.stanford.edu<mailto:bioontology-support at lists.stanford.edu>
https://mailman.stanford.edu/mailman/listinfo/bioontology-support

========================
John Graybeal
Technical Program Manager
Center for Expanded Data Annotation and Retrieval /+/ NCBO BioPortal
Stanford Center for Biomedical Informatics Research
650-736-1632
skype: graybealski


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.stanford.edu/pipermail/bioontology-support/attachments/20170301/eb0e3af6/attachment-0001.html>


More information about the bioontology-support mailing list