Bob from Mayo Clinic is here.

<mscottm> https://docs.google.com/document/d/1lKdDSb2uBBIeTEQAv2CyTHN_aVW63k9si1hmmilOMi0/edit?authkey=CJGUtcwF&hl=en_US&pli=1

Scott: How to manage pgx knowledge to harness for personalized medicine apps
... context is warfarin b/c well documented and important

Bob: -1 to semweb in title

Bosse: This title is more consistent w purpose

Scott: Nice title, and improvement
... introduction has been smoothed by Matthias
... talking re pgx being incorporated in HC practice, via product labels
... intro leads up to new product label for diff alleles of warfarin gene variants
... should explain our allele notation
... the star* notation

Matthias: In table?

Scott: Or bring it up when talking re HSGV naming scheme, LRG etc
... need to explain that we focus on SNPs rather than all genetic variants
... unambiguous IDs, round out discussion
... already in TMO we have example data that make use of SNP notation
... for particular patients. Could influence a canonical representation
... demo data vs. IDs in our own KB
... VKORC1 and CYC2C9

Matthias: Map between IDs?
... return to the problem of partial solution(?)

Scott: Don't need to figure out whole solution
... for SNPs, could have procedural call to mutalyzer
... extra effort here.

Matthias: There are diff ways of spelling out the variation
... look at dbSNP

Scott: Main dbSNP IDs were classic names; also point to HSGV names when there

<matthias_samwald> http://www.ncbi.nlm.nih.gov/projects/SNP/snp_ref.cgi?rs=104894542

try rs9923231

<mscottm> http://www.lrg-sequence.org/page.php?page=about

Discussion of whether we should talk re HVGS ids

Scott: Should try to explain how these relate to HVGS

<mscottm> http://www.hgvs.org/mutnomen/

Scott: Also good idea to explain GWAS, perhaps in the context of drug development
... important to bring up that a SNP is an assertion based on statistical evidence
... slightly different kind of knowledge than what people normally represent
... have discussed representing ethnicity
... but now how to represent hapmap, if indeed SNP is relative to geopolitical boundary
... refer to pop, plus refer to boundary
... refer to ethnicity when talking SNPs always?

Matthias: Must pick the reference population
... hapmap mostly defined by geography
... guess we should try to focus on hapmap pops at the moment
... Michel already has some class to define SNPs
... hapmap pop might not be relative to some race

Scott: Distiinguished from ethnicity?

Matthias: Ethnicity irrelevant from genetic point of view; strictly culturally defined

ericP: But good correlation for certain genotypes?

Scott: Ethnicity is described in context of warfarin SNPs
... warfarin lit refers to Asians, blacks, Caucasians, etc
... we should probably change the race class to something more palatable?
... Let's keep this in mind for update!

BobF: Question of how to define race/ethnicity?

Scott: Choosing a label; more just a problem of what we're going to call it
... genes related to categories of ancestry

BobF: Race/ethnicity have strict defn relative to OMG in US, for collecting clinical trial info
... there is a clear distinction about what is race, what is ethnicity
... Had email exchange w Michel in May

(I remember, but couldn't find you either :-)

Scott: Try to fill out paper; done by EO next week
... competency questions

<BobF> US National Cancer Institute standards for race and ethnicity: https://cabig.nci.nih.gov/workspaces/VCDE/Data_Standards/CDC_Race_Eth.zip

Matthias: pharmGKB has been converted by Tim and by Michel
... should collect all the datasets on wiki for transparency

<BobF> Ethnicity data element: https://cdebrowser.nci.nih.gov/CDEBrowser/search?elementDetails=9&FirstTimer=0&PageId=ElementDetailsGroup&publicId=2192217&version=2.0

<BobF> Race data element: https://cdebrowser.nci.nih.gov/CDEBrowser/search?elementDetails=9&FirstTimer=0&PageId=ElementDetailsGroup&publicId=2192199&version=1.0

Scott: One view is that we have several diff dbs, including text-mined data
... showcase data sources, as well as bring in others?

Matthias: But do not have concensus on how to represent variations in TMO
... representation of large amount of data w all sorts of restrictions

Scott: Summarize here, to put out on pgx list

<BobF> OMB directive for race/ethnicity (see appendix 1): http://www.whitehouse.gov/omb/fedreg_directive_15 (this may be too restrictive for the purposes of this paper, but could serve as a starting point)

Elgar: Strongly +1 support putting on wiki all the sources
... one sentence summary of each.
... who is converting what?

Current version:

https://docs.google.com/document/d/1I9xyVKhO9wG7My2fWu9S-wMwWZxG6PkRc98kOxUqH-o/edit?hl=en_US

Matthias: Redundancy for dbSNP and pharmGKB

Elgar: Relevant DBs, are all sources listed in table?

Matthias: Not able to convert all of them certainly at the moment

Elgar: +1

Scott: pharmGKB conversions: Be practical here
... Michel wrote script; review w Tim and Joanne
... which one may not be important for paper; need not use one or the other
... ericP has done a lot of mapping in the KB
... to map RDF consistent w TMKB
... maybe also use SWObjects to get access
... we don't even need to talk about federation; automatically routed to different stores

ericP: Different mappings to RDF; might have use case discussion to resolve
... also suggestion to send off to google discussion list; Would prefer to push to HCLS list

Scott: Trying to wrap up rather than unwrap it

Bob: -1 here

Scott: Part of issue is how to use a reasoner and how we would do that
... Matthias points out that reasoner would require separate section
... Michel has talked re ethnicity via restrictions in owl
... something convincing for reasoning would be good, not necessary

(BobF - would you please spell your name here?)

BobF: Would like to get a better understanding w/o distracting.

Scott: Jump in!

<ericP> matthias_samwald, i don't mean to complicate modeling decisions et. al. if you'd rather discuss in smaller setting, i understand

ericP: Goal is the modeling. Need to go into ID issues
... inferencing is something we do only when motivated by use cases

BobF: How does work overlap w SO, etc?

ericP: Would need to do homework.

(Michel would be the expert in this area.)

Scott: Broader sense, lots of efforts to describe genotypes, phenotypes w ontologies
... Michel has pointed out foundational difficulties w SO
... pgx ontologies, there aren't really any around (yet)

BobF: pharmGKB started at one time work on an ontology

(breaking up)

BobF: very interested in learning

hello??

(muting confusion here :-)

BobF: Want to learn more about overlap w other work

(echo now)

BobF: PGRN pharmacogenomics research network

<mscottm> http://pgrn.org

<BobF> http://pgrn.org/display/pgrnwebsite/PHONT+Profile

<epichler> need to run - on vacation for the next 2 weeks - see you all again in August - e2.

Scott: PGRN related directly to pharmGKB

BobF: PGRN is parent project, funded to be research site
... since then pharmGKB is funded separately but still closely associated

<mscottm> mscottmarshall@gmail.com

<ericP> cheers all

rssagent, draft minutes

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.


WARNING: No "Present: ... " found!
Possibly Present: Bob BobF BobP Bob_ Bosse Elgar Matthias Scott bbalsa epichler ericP hcls2 https joined matthias_samwald mscottm
You can indicate people for the Present list like this:
        <dbooth> Present: dbooth jonathan mary
        <dbooth> Present+ amy


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: 14 Jul 2011
Guessing minutes URL: http://www.w3.org/2011/07/14-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