Re: Fwd: Nature Publishing Group Linked Data Platform

Hi Tony,

On Thu, Apr 19, 2012 at 8:27 AM, Hammond, Tony <t.hammond@nature.com> wrote:

> Hi Michel:
>
> Just wanted to respond to your earlier post [1] about NPG's Linked Data
> Platform. (And apologies for not responding sooner - I only came across
> this
> now.)
>
> > unfortunately, after a cursory look ( hope i'm wrong) - i don't think the
> > data links into anything on the semantic web... (mesh terms are literals,
> > pmids are in NPG's namespace with no links to identifiers.org, etc)
>
> I've listed below (at end of message) a given NPG :Record object which has
> these features:
>
>    1. Is typed and identified in the npg: namespace
>

yes this is good.

>
>    2. Has links to a PubMed record:
>
>    a) Uses one form of URI used by the owner:
>
>    <http://pubmed.org/20436485>
>

i wasn't even aware of this url pattern


>
>    b) Uses a second form of URI which is not dereferenceable but is a
> "standard" name (covered by RFC - and IANA):
>
>    <info:pmid/20436485>
>
>    3. Has MeSH descriptor/qualifier terms in text form (as used by PubMed)
>
> Some comments:
>
>    1. The object is from a data package used within our realm - and so is
> consistently represented as such. (All our objects have npg: types
> alongside
> other types as appropriate.)
>
>    2. As far as I am aware this data does not have canonical RDF
> representations from the ontology owner - NLM.  We have used a URI form
> that
> is used by PubMed although it is not clear if that is intended as an
> address
> rather than a name:
>
>    <http://pubmed.org/20436485>
>
> We could have selected this form - with similar caveats:
>
>    <http://www.ncbi.nlm.nih.gov/pubmed/20436485>
>
> We thought the first form was more acceptable but are ready to be
> corrected.
>
>    3. We have used an explicit name form URI:
>
>    <info:pmid/20436485>
>
>    This is not an HTTP URI - and hence will be frowned on by Liked Data
> purists.


it's not about LD purism - it's whether other people also identify the data
in a way that enables resource integration (no need to flame the fire :)  .
I've argued in the past that such identifiers are perfectly acceptable,
provided we had a common or shared format for that. We, with others, have
built identifiers.org to fulfill that role

http://identifiers.org/pubmed/20436485

It would be great to have you join this effort.


> But is is a valid RDF URI and as such functions as a unique linking
> point - albeit without offering any additional information. (One step up
> from a blank node, and two steps down from an HTTP named node.)
>
>    4. So, should we also have added one of these forms below? And if so,
> which is to be preferred? And are there others?
>
>    <http://identifiers.org/pubmed/20436485>
>    <http://bio2rdf.org/pubmed:20436485>
>    <http://purl.org/commons/record/pmid/20436485>
>
>
It's up to you which you add in. Clearly, there is a burden in maintaining
outward links. Identifiers.org provides a set of services so that you get
other links:

http://www.ebi.ac.uk/miriam/main/mdb?section=ws_help




>    5. As for MESH terms we would be willing to take some guidance. Both
> Bio2RDF (and the earlier Neurocommons work) provided URIs for descriptors
> and  qualifiers:
>
>    <http://bio2rdf.org/mesh:D005602>
>    <http://bio2rdf.org/mesh:D005602Q000208>
>    <http://purl.org/commons/record/mesh/D005602>
>
>
Bio2RDF is a safe bet - we're updating, maintaining and expanding this
service as we speak.




>    Again, which to pick?
>
>    6. It is a pity though that the ontology owner has not (to my knowledge)
> provided some guidance for how their ontology may be used within an RDF
> context.
>

which ontology owner?


>
> We are still learning and looking to improve the stability and reach of the
> triples we're making available and eager to get any helpful feedback.
>

looks good, keep it up!

m.


>
> Cheers,
>
> Tony
>
>
> [1]
> http://lists.w3.org/Archives/Public/public-semweb-lifesci/2012Apr/0023.html
>
> ==
> @prefix dc:       <http://purl.org/dc/elements/1.1/> .
> @prefix rdf:     <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
>
> <http://ns.nature.com/records/pmid-20436485>
>      a       <http://ns.nature.com/terms/Record> ;
>      <http://www.w3.org/2000/01/rdf-schema#label>
>              "20436485" ;
>      dc:identifier <http://pubmed.org/20436485> , <info:pmid/20436485> ;
>      dc:subject "Terminal Repeat Sequences" , "Lymphoma/genetics" , ... .
> ==
>
>
>
>
> ********************************************************************************
> DISCLAIMER: This e-mail is confidential and should not be used by anyone
> who is
> not the original intended recipient. If you have received this e-mail in
> error
> please inform the sender and delete it from your mailbox or any other
> storage
> mechanism. Neither Macmillan Publishers Limited nor any of its agents
> accept
> liability for any statements made which are clearly the sender's own and
> not
> expressly made on behalf of Macmillan Publishers Limited or one of its
> agents.
> Please note that neither Macmillan Publishers Limited nor any of its agents
> accept any responsibility for viruses that may be contained in this e-mail
> or
> its attachments and it is your responsibility to scan the e-mail and
> attachments (if any). No contracts may be concluded on behalf of Macmillan
> Publishers Limited or its agents by means of e-mail communication.
> Macmillan
> Publishers Limited Registered in England and Wales with registered number
> 785998
> Registered Office Brunel Road, Houndmills, Basingstoke RG21 6XS
>
> ********************************************************************************
>
>
>


-- 
Michel Dumontier
Associate Professor of Bioinformatics, Carleton University
Chair, W3C Semantic Web for Health Care and the Life Sciences Interest Group
http://dumontierlab.com