W3C home > Mailing lists > Public > public-semweb-lifesci@w3.org > April 2010

RE: ontologies for BioRDF and LODD datasets

From: Michel_Dumontier <Michel_Dumontier@carleton.ca>
Date: Thu, 22 Apr 2010 11:41:19 -0400
To: François Scharffe <francois.scharffe@inria.fr>, public-semweb-lifesci hcls <public-semweb-lifesci@w3.org>
Message-ID: <E1784B0107E5634C8997868083EDE7805CCDAF4518@CCSMBX10.CUNET.CARLETON.CA>
Hi François,

 You can imagine that Linking Open Data (LOD) is the first generation of making *open data* available with semantic web technologies. It takes significant amounts of time and effort to determine whether "drug" is the same or different concept in DrugBank, PharmGKB, Dailymed, ChEBI, etc -> and this is what new ontology-based efforts like the Translational Medicine Ontology [1] attempt to address with ontology + data + mappings [2]. So LOD efforts like LODD or Bio2RDF are generally more focused on getting the data as part of the LOD cloud [3], with which new efforts can semantically integrate these.



[1] http://esw.w3.org/HCLSIG/PharmaOntology 
[2] http://esw.w3.org/HCLSIG/PharmaOntology/TMKB
[3] http://linkeddata.org/ 

-----Original Message-----
From: public-semweb-lifesci-request@w3.org [mailto:public-semweb-lifesci-request@w3.org] On Behalf Of François Scharffe
Sent: Thursday, April 22, 2010 7:45 AM
To: public-semweb-lifesci hcls
Subject: ontologies for BioRDF and LODD datasets


I've recently played around with BioRDF and LODD datasets. I must say 
I'm impressed by the number of datasets available. It's good to see how 
the linked data can lead to interesting queries allowing to discover new 
information. I particularly like the TCM examples [1].

I have a concern regarding the vocabularies used to published datasets.
On one side it seems there are plenty of biomedical ontologies like the 
ones on the bioportal, but on the other side there are many datasets 
that each use its own native schema without referring to these ontologies.
Is there for example a reason why 'drugs' concepts in Dailymed and 
Drugbank are different ? Why not reusing an existing ontology describing 
drugs ? Is there actually a set of datasets reusing the same ontology/ies ?

I'm actually working on ontology alignment so I'd be glad to help in the 
aligning task. I must however admit that I'm not very knowledgeable in 
the biomediacal domain.


[1] http://code.google.com/p/junsbriefcase/wiki/RDFTCMData
Received on Thursday, 22 April 2010 15:41:58 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:52:43 UTC