- From: Alasdair J G Gray <A.Gray@cs.man.ac.uk>
- Date: Tue, 18 Dec 2012 17:32:41 +0000
- To: Michel Dumontier <michel.dumontier@gmail.com>
- Cc: Peter Ansell <ansell.peter@gmail.com>, Umutcan ŞİMŞEK <s.umutcan@gmail.com>, public-semweb-lifesci@w3.org, Stian Soiland-Reyes <soiland-reyes@cs.manchester.ac.uk>, Paul Groth <p.t.groth@vu.nl>
- Message-Id: <C402EF4C-BDCD-4204-9CE6-76C08C8F3F50@cs.man.ac.uk>
Hi Michel, On 13 Dec 2012, at 04:39, Michel Dumontier <michel.dumontier@gmail.com> wrote: > Hi Alasdair, > Our library does have support for source data versions, but we currently specify this with the "version" predicate from biositemaps > > http://www.ncbcs.org/biositemaps/formatDescription.html > I still need to look into BioSiteMaps and their relationship with the VoID work we have done in Open PHACTS. My preference so far has been for VoID since it is generic, i.e. not specific to bio-informatics, and that the descriptions can be tied to the data. > I don't have any preference, but i would perhaps encourage us to consolidate provenance information with the prov-o effort. > > moreover, http://purl.org/pav/version does't answer - and I would like to see an RDF representation. > I will get Stian to look into making the resource resolvable, but the whole owl file is available from http://purl.org/pav/. There is currently an effort to align pav with prov, so that will come in due course. > I added versioning, where possible, to Bio2RDF Release 3. > :) Alasdair > https://github.com/bio2rdf/bio2rdf-scripts/issues/milestones > > m. > > > On Wed, Dec 12, 2012 at 6:12 AM, Alasdair J G Gray <A.Gray@cs.man.ac.uk> wrote: > Thanks for the link. Great to see you using VoID to describe the data. > > I'd argue for having a triple capturing the explicit version number of the DrugBank dataset using a triple like the following. My reason is that we encountered a lot of problems in Open PHACTS tracing what versions of datasets were being used when the only information we had to go on was the conversion date. > <http://bio2rdf.org/bio2rdf_dataset:drugbank> <http://purl.org/pav/version> "3.0" . > > I note that you describe both the RDF version of DrugBank and the original DrugBank data as a void:Dataset. My understanding of the class definition is that the dataset should be captured in RDF [1]. For this reason, in our Open PHACTS work we would tend to use the dctype:Dataset class to define the original DrugBank data [2]. > > Alasdair > > [1] http://vocab.deri.ie/void#Dataset > [2] http://dublincore.org/documents/2012/06/14/dcmi-terms/?v=dcmitype#Dataset > > On 12 Dec 2012, at 10:40, Michel Dumontier <michel.dumontier@gmail.com> wrote: > >> Hi, >> The parser works on the v3 XML files. We don't currently specify the source dataset at that level of detail - here's the link to the provenance: >> >> http://download.bio2rdf.org/release/2/drugbank/bio2rdf-drugbank-20121008.nt >> >> m. >> >> On Wed, Dec 12, 2012 at 4:53 AM, Alasdair J G Gray <A.Gray@cs.man.ac.uk> wrote: >> Is there a metadata description of the data set capturing this kind of information? >> >> >> >> -- >> Michel Dumontier >> Associate Professor of Bioinformatics, Carleton University >> Chair, W3C Semantic Web for Health Care and the Life Sciences Interest Group >> http://dumontierlab.com >> > > Dr Alasdair J G Gray > Research Associate > Alasdair.Gray@manchester.ac.uk > +44 161 275 0145 > > http://www.cs.man.ac.uk/~graya/ > > Please consider the environment before printing this email. > > > > > -- > Michel Dumontier > Associate Professor of Bioinformatics, Carleton University > Chair, W3C Semantic Web for Health Care and the Life Sciences Interest Group > http://dumontierlab.com > Dr Alasdair J G Gray Research Associate Alasdair.Gray@manchester.ac.uk +44 161 275 0145 http://www.cs.man.ac.uk/~graya/ Please consider the environment before printing this email.
Received on Tuesday, 18 December 2012 17:33:08 UTC