W3C home > Mailing lists > Public > www-rdf-interest@w3.org > March 2004

RE: Using RDF to describe biological taxonomy.

From: Chris Catton <chris.catton@zoology.oxford.ac.uk>
Date: Mon, 29 Mar 2004 15:44:51 +0100
To: "'Graham Klyne'" <GK@ninebynine.org>, "'Tim Nowaczyk'" <zimage@cs.wisc.edu>, <www-rdf-interest@w3.org>
Cc: "David Shotton" <david.shotton@zoology.oxford.ac.uk>
Message-ID: <000701c4159c$648a8e60$97184381@chios>

Hi there,
Graham is quite right that we are doing work with DAML+OIL, but I don't
think anything we have done yet is directly relevant to the question
posed here.  Basically we are building an ontology of (mainly
microscopy) image description and annotation - so at the simplest level
we have triples like 
Image : has_subject : species

The modelling Eric posted seems eminently sensible, but we simply don't
address the question. We store the NCBI taxonomy in a nested-set
representation in a postgres database. We've done this simply for speed
of retrieval - I'd be interested to know how an rdf triple store
performs - eg in response to a query like 'get all species of primates'.


We identify the species with a urn that uses the ncbi taxonomy id, and
currently that's about it, though we do want to start (very soon) doing
some more interesting things like using the taxonomy to help search for
images. So for instance, in response to the query 'find me all images of
primates' the system should return images containing representations of
apes, monkeys etc.

We also envisage returning rdf graphs along with the images - harvesting
additional information from a number of different sources to build them.
Quite how this will be achieved we don't know yet - this is after all a
research project.

I'd be very happy to continue this discussion with anyone interested -
though it may be better done off list since I guess it's of only
marginal interest to most here?


Chris Catton
BioImage Database Development Manager
Department of Zoology
University of Oxford
OX1 3PS
 
Tel: +44 (0) 1865 281993
email: chris.catton@zoology.oxford.ac.uk
web site: www.bioimage.org


> -----Original Message-----
> From: www-rdf-interest-request@w3.org [mailto:www-rdf-interest-
> request@w3.org] On Behalf Of Graham Klyne
> Sent: 29 March 2004 12:22
> To: Tim Nowaczyk; www-rdf-interest@w3.org
> Subject: Re: Using RDF to describe biological taxonomy.
> 
> 
> An indirect answer:  unfortunately, I'm offline so I can't track down
the
> links.
> 
> I vaguely recall that the zoology department at Oxford University are
> doing
> some work with RDF and/or DAML/OWL.  I haven't a clue if its relevant
to
> your question, but seems to me it's worth tracking down/checking out.
> 
> [later] URI:  http://www.bioimage.org/
> 
> #g
> --
> 
> At 22:26 26/03/04 -0600, Tim Nowaczyk wrote:
> 
> >I've set this up, and I'd like your comments.  I've used the ITIS
> >database along with some perl to describe biological taxonomy
(Kingdom,
> >Phylum, etc..)  in the style of wordnet.
> >
> >Homo Sapiens (species)
> >http://66.168.24.195:8081/~tim/taxonomy/0.1/tsn/180092
> >
> >Primates (Order)
> >http://66.168.24.195:8081/~tim/taxonomy/0.1/tsn/180089
> >
> >I'd like to get this to production quality.  I wanted avoid creating
my
> >own vocabulary, which is why I am using rdfs:comment for the rank
> >and rdfs:title for the latin name.  The more I think about it, the
more
> >I'm convicted to create a vocabulary.
> >
> >Before I go any further, I'd like your input.  Is there a better way
of
> >doing this?  Is this a stupid idea that noone will ever use?  I'm new
to
> >RDF, so I need your comments.
> >
> >Thanks for your time,
> >Tim Nowaczyk
> >
> >--
> >----------Tim
Nowaczyk----------zimage@upl.cs.wisc.edu------------------
> >| GPG fingerprint = 1612 3E41 D649 63ED 222D  EB27 E84D 4274 392C
A50A
> 
> ------------
> Graham Klyne
> For email:
> http://www.ninebynine.org/#Contact
> 
Received on Monday, 29 March 2004 09:45:10 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 18 February 2014 13:20:07 UTC