RE: Chemistry and the Semantic Web

I was just looking at Oracle's Spatial Network Data Model this morning
and wondering if that might be a good place to start. This may be overly
simplistic, but the NDM Schema has two components, Meta Data (Name, Type
and the Table info) and Tables for Nodes, Link's and Paths.  
 
Would take more time to think this through, but perhaps it something you
may want to take a look at.
 
Joanne


  _____  

	From: public-semweb-lifesci-request@w3.org
[mailto:public-semweb-lifesci-request@w3.org] On Behalf Of
Eric.Neumann@aventis.com
	Sent: Tuesday, June 29, 2004 5:15 PM
	To: public-semweb-lifesci@w3.org
	Subject: Chemistry and the Semantic Web
	
	



	In looking at some areas outside of the scope of traditional
bioinformatics (but still part of drug discovery), it would appear that
the Chemistry Domain could greatly benefit from Semantic Web approaches.
For instance, it might be useful to somehow allow scientists to annotate
select groups of chemical structures using interconnecting relations
that have specific significance to a drug discovery project. 

	Example: a set of azaindoles may show CDK2 toxicity if they
contain a polar group at position 3 on the indole group. A mechanism of
action may even be proposed (using RDF) for this toxicity and associated
with the non-selective binding to a pocket on CDK2. Alternatively, some
side groups could be found to dramatically affect Pharmacokinetics (PK),
so this would be great to RDF-link-in as well.

	Question: How would one apply RDF for such cases? Would one use
CML (chemical markup language) to describe the chemical structure and
have an RDF statement refer to part of that doc via XPath/XPointers? How
about other structural formats like SMILE and CHUCKLES? Would the
documents be referenced using an LSID mechanism? Could this become the
basis for allowing research findings around chemistry and assays to
become consolidated as part of a R&D knowledge base?

	This is a big issue here at Aventis with many projects. Linking
mechansisms and interactions to chem structures would be significant for
enhancing innovation and productivity. I think this builds on applying
LSIDs to chemical structures (via CAS, etc), but goes beyond on how to
represent the interpretations of assays and models of action as part of
drug development.

	Eric 


	Eric Neumann, Ph.D. 

	Global Head of Knowledge Management 
	Aventis - DI&A 
	Tel:   908-231-3510 
	Fax:  908-231-3307 
	Eric.Neumann@Aventis.com