[open-science] HACKATHON-Semantic Web Identifiers for bioscience

Fri Dec 2 19:46:00 UTC 2011

Puneet --

Just some relevant pointers... I'm sure you've already come across the TDWG
(Taxonomic Databases Working Group) standards for publishing biological
taxonomies, names, and concepts, such as the Taxonomic Concept Transfer
Schema (TCS; http://www.tdwg.org/standards/117/).  Also relevant are
implementations of semantic concept linking at http://taxonconcept.org, as
well as more traditional links to taxon names through the Global Names
Index (GNI; http://www.globalnames.org/)?

PS I was involved in discussions about paleodb.org when Alroy first
launched it at NCEAS; I'm happy to see its had such longevity and is
supporting such an effective open data community.

Matt Jones
NCEAS, UC Santa Barbara

On Fri, Dec 2, 2011 at 10:21 AM, Puneet Kishor <punkish at eidesis.org> wrote:

> My apologies if this is tangential, but I came late to this conversation
> on taxonomy and am trying to reconstruct the thread.
>
> I work on the team that manages paleodb.org which maintains the taxonomic
> information of fossils. I am working on a new architecture for an
> application that will offer paleodb's taxonomic information RESTfully and
> compliant with all the current best practices.
>
> Perhaps a separate list would be worthwhile to discuss/share ideas on
> creating taxonomy applications.
>
> On Dec 2, 2011, at 11:46 AM, Peter Murray-Rust wrote:
>
> > Jenny has highlighted that we shall be using this list to discuss the
> > hackathon. I suggest we use a separate title for each thread, prefaced by
> > HACKATHON-
> >
> > My problem is how to create indentiers for (say) viruses. If you look at
> > Wikipedia it doesn't give IDs. But I then discovered (by chance) taxid:
> > which gives numbers. But the pages don't give static URIs (they contain
> > cgi). I am cutting and pasing the discussion (and I shal;l refer any
> others
> > to here.
> >
> > Jerven Bollema
> >
> >
> > Hi Peter, All,
> >
> > All taxons in the UniProt taxonomy can be found via (
> > http://purl.uniprot.org/taxonomy/10305). This is synchronized with the
> NCBI
> > taxonomy and is the same in the public version (release delta excepted).
> > Some limited NCBI taxonomy curation happens at the Swiss-Prot group which
> > also does the UniProt rdf work (Guess where I work ;).
> >
> > In this case you actually have an link in rdf from the herpes virus to
> its
> > hosts. The proteins it encodes (might not be all for each virus isolate
> > e.g. in this case only a single virion membrane protein is known) and
> links
> > to relevant papers as well as related virion proteins.
> > Will love to show you all how you can get this data in RDF and work with
> it
> > using SPARQL.
> >
> > Regards,
> > Jerven
> > =============================================
> >
> > M. Scott Marshall
> > show details 11:36 AM (6 hours ago)
> >
> > Dear Peter and Jerven,
> >
> > Nice blog with dawg and frog!
> >
> > Thanks for the answer Jerven. I'm looking forward to this.
> >
> > Not wanting to start (too much) commotion but also bumped into this
> > for taxons: http://rs.tdwg.org/dwc/index.htm
> > From recent conversations stemming from the biohackathon in Kyoto, I
> > surmise that the NCBI taxonomy, presumably with a bias toward human
> > and model organisms, as well as its adoption by Uniprot, should be the
> > preferred choice for expressing taxon info in the context of
> > biomedical knowledge?
> >
> > Next question (thinking that I know the answer) is how to integrate
> > datasets that use one of each: NCBI and Darwin? Of course, we can
> > shelve this and come back to it some other year, if we want to dig
> > into something more specifically biomedical. The issue will eventually
> > come back to haunt us in any case. For example, with chemical
> > identifiers..
> >
> > Also, is taxon out of scope for Identifier.org ?
> >
> > Cheers,
> > Scott
> >
> > ===================================
> > [Jerven]
> > Hi All,
> >
> > Don't wish to spam the mailing list about which taxonomy to use. However
> if
> > you actually look at the darwin and uniprot taxonomy "schema" then they
> are
> > very similar. And even in the taxonomy world it doesn't have that many
> > controversies.
> >
> > Its the instances that get hairy.
> > i.e. is it
> >
> > Dugu is a rodentia
> > <purl.uniprot.org/taxonomy/10160> rdfs:subClassOf <
> > purl.uniprot.org/taxonomy/9989>
> > or
> > Dugu is a Caviomorpha
> > <purl.uniprot.org/taxonomy/10160> rdfs:subClassOf <
> > http://dbpedia.org/resource/Caviomorpha>
> >
> > Which gets taxonomist all exited :)
> >
> > Mapping schemas is easy to do here. Its mapping instances that get the
> > feuds started :D
> >
> > Regards,
> > Jerven
> >
> > ============================================
> > PMR comment - this isn't spam, it's science!
> > ============================================
> > PMR - thanks for this. If we can make progress on identifiers it makes
> *me*
> > happy!
> > --
> > Peter Murray-Rust
> > Reader in Molecular Informatics
> > Unilever Centre, Dep. Of Chemistry
> > University of Cambridge
> > CB2 1EW, UK
> > +44-1223-763069
> > _______________________________________________
> > open-science mailing list
> > open-science at lists.okfn.org
> > http://lists.okfn.org/mailman/listinfo/open-science
>
>
>
> --
> Puneet Kishor http://punkish.org
> science http://earth-base.org
> advocacy http://creativecommons.org
>
>
>
> _______________________________________________
> open-science mailing list
> open-science at lists.okfn.org
> http://lists.okfn.org/mailman/listinfo/open-science
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.okfn.org/pipermail/open-science/attachments/20111202/1670737b/attachment-0001.html>