<micheldumontier> on my way

<adrien> HI everyone - unfortunately I have to leave at 18h20

<micheldumontier> we'll make it short

<ericP> micheldumontier, will be there in 5 mins

<mscottm> I sent a note to Fred Whipple to ask him to join us since he was involved in the formative months.

scribenick bobP

Michel: discussing large sets of interconnected dbs...
... next step, need to demonstrate the integration
... should show URIs for docs etc
... motivation for the owl section
... how to communicate the essence of the tech, so readers have a grip

Adrien: Lots of data expressed in natural language
... (missed some here..)

Scott: Chris Baker has reference here..

<mscottm> http://www.biomedcentral.com/1471-2164/11/S4/S24

Scott: Mutation impacts, from text mining
... closely aligned with what we are describing

<adrien> I have to run - have a good meeting

<ericP> i'll cover for you

Michel: Figure for cds clincical decision support

<adrien> tks

Michel: table has several levels - (I see it)

BobF: Level 1, before variant can be used in pheno-geno, needs to be validated etc
... there is work around clean-up and nail-down before clinical application

Michel: Level 1 is basic research; don't know yet re importance
... don't yet know relationship to pheno

BobF: Reading tightly; if includes assay and validation then it's OK
... validation that it's real, not an artifact

Michel: Level 1 is appropriate for validation too

BobF: Today most of SNPs go thru one level of validation

Michel: RHS has orders of magnitude, can we give reference?

BobF: Thought about, not sure so easy
... good for graph representation of funnel shape rather than exact numbers

Scott: Funnel shape gets the idea across

BobF: Numbers might tie this to time

<ericP> funnel shape will be familar to folks who've seen popular representations of the attrition of drug leads in a development pipeline

Michel: Need a bang-for-the-buck ending

Scott: Need one more bullet at least

BobF: Title is focussed on infra and tech, and how they facilitate
... concluding portion should help reinforce that
... many different components: datasets, onts, etc
... bring all of threads together; real advance in brining all these resources to bear
... (this is too good to miss - write it out!)

Michel: Should end in a different place?

Scott: Need a punch line. Everybody is after mapping geno-pheno
... look at that as a bottleneck; When we have these mappings then easy to share
... one of the promises of semweb, describe sets of phenos in commonly accepted form
... and access snps in global namespace, then harness to make statements understood here to China

BobF: Many of datasources have evolved over time, become internally standarized
... dbSNP great effort to make unique IDs
... single term for concepts. But this evolution is still in silos!
... semweb can start us to more seamlessly normalize across silos

Elgar: Main concern, why do all of this, what is the real advantage of owl, need example?
... meaning of interlinking data. Matthias has shown how to make queries across dbs
... section on owl inference, using genetic info to stratify patients
... would be nice to show that w owl can be easy
... under figure on integration, using the functionalities of owl...
... nice to show how things are much easier

ericP: Here's how traditional query, vs here's how to do w owl

<mscottm> immediate advantages to SPARQL endpoint: subsumption reasoning, federation

ericP: can mock up SQL w a couple of hospital dbs

Michel: Patient has some disease, affects some morphological part

Elgar: Agree, but want one simple example of superiority.
... Under figure by Matthias, we can conceptualize...as a pipeline...
... don't quite see the analogy here.

<micheldumontier> https://docs.google.com/document/d/1bsJ3YWuH1k3gdXtSN2PXfmUIavtR-hDXhHMn0bKB5pk/edit?hl=en_US

DaveHau: will look at report

Scott: Deadline tomorrow :)
... Need to be clear about benefits of clinical practice vs clinical research
... Fred Whipple was pointing this out.

<micheldumontier> dave: i need you to request access to the document

Scott: info retrieval, hypothesis testing need to be kept separate
... we are saying that both will benefit.
... clinical software should alert to dose-response w allele
... too difficult for doctor to do like a researcher
... this is a benefit of semweb infra (to automate this?)

Dave Hau: There is a feedback between clinical practice and research(?)

<mscottm> The notion of feedback (Dave mentioned) is probably worth mentioning in the 'future prospective' section.

This is scribe.perl Revision: 1.136  of Date: 2011/05/12 12:01:43  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

No ScribeNick specified.  Guessing ScribeNick: bobP
Inferring Scribes: bobP

WARNING: No "Topic:" lines found.

Default Present: Bob_Powers, Scott_Bauer, +31.62.427.aaaa, +33.3.83.59.aabb, +1.781.431.aacc, +1.781.431.aadd, +1.301.443.aaee, EricP
Present: Bob_Powers Scott_Bauer +31.62.427.aaaa +33.3.83.59.aabb +1.781.431.aacc +1.781.431.aadd +1.301.443.aaee EricP

WARNING: No meeting title found!
You should specify the meeting title like this:
<dbooth> Meeting: Weekly Baking Club Meeting


WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth

Got date from IRC log name: 29 Sep 2011
Guessing minutes URL: http://www.w3.org/2011/09/29-hcls2-minutes.html
People with action items: 

WARNING: No "Topic: ..." lines found!  
Resulting HTML may have an empty (invalid) <ol>...</ol>.

Explanation: "Topic: ..." lines are used to indicate the start of 
new discussion topics or agenda items, such as:
<dbooth> Topic: Review of Amy's report