Re: Banff demo

Hi Ivan,

Here's the revision of the abstract that was drawn directly from the  
Banff presentation PDFs:



posted by Bill Bug  on 5/12 as a summary both of Don & Matthias copy  
and the Banff presentations  -William Bug 5/12/07 10:22 PM

Accelerating integrative neuroscience research through Semantic Web  
technology
The enormous scientific and clinical progress hastened via large- 
scale efforts to systematically express public bio-molecular sequence  
and structure databases according to community-shared syntax has been  
considerably enhanced in the last decade via systematic, explicit  
association of semantic expressions defining relevant, meaningful  
properties and relations (e.g., the molecular function, cellular  
component, and biological function described by The Gene Ontology).   
This is part of a larger effort to apply consistent semantic  
annotations to biomedical information from a variety of sources  
(public data repositories, scientific literature corpi, clinical  
trials and medical reports, etc.). The goal is to facilitate  
semantically-driven data integration and queries, thus avoiding  
duplication of research effort, uncovering deep, meaningful  
correlations across a broad spectrum of experiments, and making more  
effective use of basic and clinical research.  In the treatment of  
complex human diseases, accelerating such broad-scoped biomedical  
knowledge discovery is most urgently needed, especially to alleviate  
the enormous damage and suffering caused both to individuals and  
society by the myriad of neurodegerative diseases such as  
Alzhiemer's, Parkinson's, and MS.  Here the W3C Semantic Web Health  
Care/Life Science Interest Group provides a focused demonstration of  
how to specifically effect such gains by using W3C-sponsored Semantic  
Web technology (SemWebTech) - RDF, OWL, and the cornucopia of robust  
tools built on these core formalisms.  We demonstrate how SemWebTech  
specifically excels at: fusing data across scientific disciplines;  
enhancing specificity of evidentiary provenance; re-combining  
original data in novel ways via inference and querying at varying  
granularity levels; extensively characterizing data inconsistencies;  
greatly extending automation of these tasks.  A SemWebTech  
application typically begins with careful modeling of the underlying  
biological reality represented as simple subject-predicate-object  
statements (RDF triples) via unambiguous, network-accessible  
identifiers (URIs). These triple stores (or RDF triple  
representations of original data repositories) are then enhanced via  
reasoner inferencing, which can also extend the complexity and  
expressivity of queries.  SemWebTech queries are resolved via SPARQL,  
and RDF-driven visualization tools both simplify result presentation  
and promote uncovering complex relations.  In this specific demo we  
focus on exploring the molecular pathology of amyloid-driven damage  
in Alzheimer's disease.  We show how SemWebTech can specifically aid  
in exploring dendritic cell biology seeking candidate genes,  
proteins, molecular functions, and cellular components effected by  
maturation of amyloid placques in dendrite-rich neuropil.  We also  
demonstrate identification of potential drug targets to treat AD- 
associated cortical Pyramidal cell pathophysiology using a relevant  
domain-restricted ontology and an RDF triple representation of  
related literature and bio-molecular data repositories.  We  
demonstrate a mashup combing queries results againsts an RDF triple  
representation of descriptive information from the Allan Brain Atlas  
with the Google Maps interface can provide a very flexible,  
alternative query and visualization framework to the ABA's 20,000  
gene-specific histologically imaged C57Bl/6J mouse brains.  Finally,  
we use the Lisp Semantic Web (LSW) tool for real-time interactive  
queries exploring a 200 Megatriple repository of MeSH annotated  
literature.  Future work will extend this demonstration by adding OWL- 
based ontologies describing several well-known neuroinformatics  
repositories (e.g., SenseLab, the Brain Architecture Management  
System (BAMS), the Cell-Centered Database (CCDB), PDSP Ki database),  
linking to RDF triple views of their underlying data repositories,  
and adding de novo constructed neuroscience RDF repositories such as  
the SWAN-based Alzheimer Research Forum hypotheses collection.  We  
will use these to extensively explore APP effects on fast- 
inactivating K+channels (I.K.A) in CNS neurons - an emerging research  
focus - to uncover fundamental etiopathological mechanisms in AD.  We  
will also demonstrate use of the ABA/Google Maps mashup in other  
neuroinformatic tools - i.e., the Mouse BIRN Atlasing Tool (MBAT).



It's pretty dense.  Folks felt it was too jargon filled for a  
neuroscience community - and too narrowly focussed on the demo  
presentations themselves.  Perhaps both of those characteristics  
would make this version (or a minor edit of it to better suit it to a  
blog post and eliminate some of the non-standard shorthand - e.g.,  
SemWebTech) would fit the purpose you describe below?

Again - I would defer to Alan and Susie as the Banff presenters to  
determine to whether they believe this truly encapsulates what they  
presented - and the future directions those presentations pointed  
toward.

Cheers,
Bill

On May 16, 2007, at 5:07 AM, Ivan Herman wrote:

> I have seen the mail of Bill Bug on the abstract, and I was wondering
> whether somebody of your group could write a one-two paragraph  
> abstract
> on the demo, with pointers, that could be added to the Semantic Web
> Activity News:
>
> http://www.w3.org/2001/sw/anews
>
> this is, in fact, a blog whose rss feeds are picked up quite  
> widely. If
> you agree in a small text, Eric or Tonya should blog it on the page  
> (it
> looks better if it is published under their name and not mine)
>
> Ivan
>
> Alan Ruttenberg wrote:
>>
>> I have updated the page http://esw.w3.org/topic/HCLS/Banff2007Demo  
>> with
>> slides, pointers to the triple store etc.
>>
>> -Alan
>>
>>
>
> -- 
>
> Ivan Herman, W3C Semantic Web Activity Lead
> URL: http://www.w3.org/People/Ivan/
> PGP Key: http://www.cwi.nl/%7Eivan/AboutMe/pgpkey.html
> FOAF: http://www.ivan-herman.net/foaf.rdf



Bill Bug
Senior Research Analyst/Ontological Engineer

Laboratory for Bioimaging  & Anatomical Informatics
www.neuroterrain.org
Department of Neurobiology & Anatomy
Drexel University College of Medicine
2900 Queen Lane
Philadelphia, PA    19129
215 991 8430 (ph)
610 457 0443 (mobile)
215 843 9367 (fax)


Please Note: I now have a new email - William.Bug@DrexelMed.edu