15:58:46 <RRSAgent> RRSAgent has joined #hcls2
15:58:46 <RRSAgent> logging to http://www.w3.org/2011/04/14-hcls2-irc
16:00:43 <mscottm> mscottm has joined #hcls2
16:02:48 <matthias_samwald> matthias_samwald has joined #hcls2
16:04:12 <bobP> bobP has joined #hcls2
16:04:17 <mscottm> zakim, this is tmo
16:04:17 <Zakim> ok, mscottm; that matches SW_HCLS(TMO)11:00AM
16:04:31 <bobP> scribenick bobP
16:04:40 <mscottm> thanks bobP!
16:04:56 <michel> regrets: Elgar, Trish
16:05:42 <bbalsa> bbalsa has joined #HCLS2
16:05:51 <joanne> joanne has joined #hcls2
16:05:52 <Zakim> + +46.4.63.3.aacc
16:06:00 <joanne> hello!
16:06:34 <joanne> What's "Idler's" mean (on the right hand side of mibbit)
16:08:16 <michel> elgar presented TMO at bio-it world
16:08:54 <Zakim> + +33.3.54.95.aadd
16:09:28 <mscottm> Zakim, who is here?
16:09:28 <Zakim> On the phone I see Bob_Powers, +62427aaaa, ??P13, +1.518.276.aabb, ??P4, +46.4.63.3.aacc, +33.3.54.95.aadd
16:09:30 <Zakim> On IRC I see joanne, bbalsa, bobP, matthias_samwald, mscottm, RRSAgent, Zakim, michel, adriencoulet, ericP
16:09:40 <bobP> Michel: Discussion of paper, more detail about use case
16:09:42 <mscottm> Zakim, +33.3.54.95.aadd is adriencoulet
16:09:42 <Zakim> +adriencoulet; got it
16:09:57 <mscottm> Zakim, +62427aaaa is mscottm
16:09:57 <Zakim> +mscottm; got it
16:10:19 <bobP> ... Adrian Coulet w. sopharm; We are lucky to have Adrian participate
16:10:23 <mscottm> Zakim, +46.4.63.3.aacc is bbalsa
16:10:23 <Zakim> +bbalsa; got it
16:10:41 <adriencoulet> thank you !
16:11:02 <bobP> ... 1) warfarin 2) breast cancer; 2) let's talk about this story
16:11:49 <bobP> Scott: Talked w Daniel Rueben(?) and John Madden about bc
16:11:57 <adriencoulet> Daniel Rubin
16:12:04 <adriencoulet> ?
16:12:24 <bobP> ... lots of people understand the terminology involved
16:12:46 <bobP> ... consider oncoDX and Nanoprint(?)
16:13:14 <bobP> ... making use of molecular info about patient in order to stratify, to make better choices
16:14:13 <bobP> ... also discover clinical trials and drugs in development
16:14:24 <michel> s/Nanoprint/Mammaprint
16:14:44 <bobP> ... easily understandable to the audience; reduces costs and needless suffering
16:15:02 <michel> OncoDX: http://www.oncotypedx.com/en-US/Breast/HealthcareProfessional/ClinicalSummary.aspx
16:15:07 <bobP> ... increases patient care.  All nice goals for theoretical use cases
16:15:23 <michel> Mammaprint: http://www.agendia.com/pages/mammaprint/21.php
16:15:30 <bobP> ... note that each use case applied to millions of women
16:16:07 <bobP> ... also clinician-single patient application. Still needs to be implemented in hospital IT
16:16:30 <michel> paper: https://docs.google.com/document/d/1lKdDSb2uBBIeTEQAv2CyTHN_aVW63k9si1hmmilOMi0/edit?hl=en&authkey=CJGUtcwF#
16:16:32 <bobP> Scott: I wrote a few things in
16:17:22 <bobP> ... not sitting on bc data, but we do have some kidney transplant data.
16:17:45 <bobP> ... might be able to get some data from Mammaprint people
16:18:16 <bobP> Joanne: Data are critical; any public bc datasets?
16:18:26 <bobP> ... can bring in IO too
16:18:47 <bobP> Scott - breaking up!
16:19:32 <bobP> Scott: ... data did not have info, embedded in a giant object model
16:19:54 <bobP> ... dropped effort b/c did not want to parse thru it all.
16:20:33 <bobP> ... working w theoretical scenarios, hardest challenge is to come up w realistic gx profile
16:20:55 <bobP> ... nobody reading article will need to look at differential expression
16:21:10 <bobP> ... we are putting forth a proof-of-concept
16:21:22 <bobP> ... but, true, the more realistic the better
16:21:47 <adriencoulet> http://www.pharmgkb.org/do/serve?objId=PA443560&objCls=Disease#tabview=tab6
16:22:00 <adriencoulet> some PGx about breast neoplasms
16:22:38 <bobP> Joanne: Knows group w/in Zach's dept
16:23:01 <bobP> ... Ponolacco(?)
16:23:38 <bobP> Michel: Adrian has posted this link w. data
16:23:48 <Zakim> -adriencoulet
16:24:00 <joanne> thanks Adrien!
16:24:13 <bobP> We lost Adrian
16:24:52 <joanne> Peter Tonellato <Peter_Tonellato   at   hms.harvard.edu>
16:25:08 <Zakim> +adriencoulet
16:25:29 <bobP> Michel: Shows down to alleles
16:26:03 <bobP> ... mostly tamoxifen datasets
16:26:05 <mscottm> All: when you aren't speaking, please mute (it should help us to get rid of the echo)
16:26:47 <joanne> Web Site for Peter Tonellato's group:  http://lpm.hms.harvard.edu/
16:27:41 <bobP> Michel: #7 looks good!  Joanne agrees
16:28:09 <michel> http://www.pharmgkb.org/do/serve?objId=PA646603&objCls=PhenotypeDataset#tabview=tab2
16:28:53 <bobP> Males do get bc.
16:31:38 <bobP> Michel: Cool w. alleles, but no outcomes here. Need to read the paper to understand remission
16:32:36 <bobP> ... convert to RDF, then query?
16:33:10 <bobP> ... these are the actual data; Not been formalized to RDF yet.
16:33:29 <bobP> ... will require some significant curation. Columnn headers require some work.
16:34:09 <bobP> ... base change definitions, we will have to parse this on a case-by-case basis
16:34:37 <bobP> Matthias: commenting here
16:34:54 <bobP> Joanne will develop use cases
16:35:32 <bobP> Michel: Need to download the data and parse
16:36:04 <matthias_samwald> e.g., http://www.snpedia.com/index.php/CYP3A5 lists the allele names (*1, *5 etc.), but unfortunately not in a way that would make it show up in the current RDF version of SNPedia.
16:36:17 <bobP> Joanne: TMO or separate ontology?
16:37:00 <bobP> Matthias: Started ont, tried to align everything w TMO and other authorities
16:37:22 <bobP> ... raw at the moment. Want to use it for clinical decision support
16:38:23 <bobP> Scott: Do you have use cases? Interviewing clinician?
16:38:40 <bobP> Matthias: Blue sky research at the moment.
16:39:31 <mscottm> Univ. of Nancy
16:40:17 <bobP> Adrian: PhD pharmacogenomics text mining
16:40:49 <mscottm> Adrien: Prof at Univ of Nancy, did PhD with PharmGKB, from Mark Musen's lab (just returned from NCBO to France)
16:41:18 <mscottm> Adrien recently published work on SO-Pharm
16:41:24 <bobP> Michel: Submit 1st week in May(?)
16:41:40 <adriencoulet> http://www.loria.fr/~coulet/
16:41:52 <bobP> ... important to develop use case, as it relates to PharmGKB website
16:44:15 <bobP> Scott: Maybe hook up w Terminology call
16:46:42 <michel> https://docs.google.com/drawings/edit?id=1OaqlmiTbIAkgHM6kCV-4xbRddHJiN1YZkCWdNHvwiGI&hl=en&authkey=CIT4rcwM
16:47:11 <bobP> Michel: Look at diagram, a general model in advance of use case
16:49:15 <bobP> ... ontology would be purple boxes
16:52:21 <bobP> Matthias asking about gene variant as attribute of population and not of person
16:53:07 <bobP> Scott: Better to have sample population associated w gene frequency
16:53:30 <joanne> i have to leave now.  Would someone volunteer to contact me later and be willing to go over the diagram with me - preferably tomorrow? Michel, ideally, but if you can't someone else who understands it as well (not having generated)
16:53:35 <bobP> ... (Scott was asking here...)
16:53:47 <joanne> bye.
16:53:53 <bobP> Matthias: Arrow from freq to pop?
16:53:54 <adriencoulet> bye :)
16:54:25 <bobP> Michel: Gene variant is still part of an individual
16:54:40 <Zakim> - +1.518.276.aabb
16:54:43 <bobP> ... to assoc freq, has to be a sub-type
16:54:51 <Zakim> -adriencoulet
16:55:42 <Zakim> +adriencoulet
16:55:47 <bobP> ... variant in population is a sub-class of variant in individual(?), has more property restriction
16:56:15 <bobP> Matthias: In my view, it's not really a subclass
16:56:40 <bobP> Scott: One class of snp, diff freqs of measurement in different population
16:57:19 <bobP> Michel: But we do assign(?) these genes to specific individuals
16:58:21 <bobP> Scott: We could have a number of models; test would be how to find w owl or sparql
16:58:43 <bobP> ... demonstate the differences of how the different models work
16:59:15 <bobP> ... brca1 has certain freq in pop, but fewer known alleles which are known to be associated w cancer
16:59:35 <bobP> ... compared to snps which are known to be associated w cancer
16:59:51 <bobP> ... gwas might have millions of snps to go thru for genome
17:00:06 <bobP> ... brca measure is more like a read-out of a prognosis.
17:00:26 <bobP> ... Carry out further w. email exchange.
17:00:48 <bobP> Michel: Add this discussion to the paper
17:01:39 <Zakim> -??P4
17:01:39 <Zakim> -Bob_Powers
17:01:41 <Zakim> -bbalsa
17:01:42 <mscottm> Scott was actually saying that the BRCA1 allele is directly associated with a type of cancer.
17:01:43 <Zakim> -??P13
17:01:50 <Zakim> -adriencoulet
17:02:11 <michel> rrsagent, draft minutes
17:02:11 <RRSAgent> I have made the request to generate http://www.w3.org/2011/04/14-hcls2-minutes.html michel
17:02:18 <mscottm> whereas you might have millions of SNPs, very few SNP combinations are directly associated with a cancer outcome
17:02:19 <michel> rrsagent, make log world-visible
17:02:57 <Zakim> -mscottm
17:02:59 <Zakim> SW_HCLS(TMO)11:00AM has ended
17:03:01 <Zakim> Attendees were Bob_Powers, +1.518.276.aabb, adriencoulet, mscottm, bbalsa
18:02:47 <david_r_newman> david_r_newman has joined #hcls2
18:18:11 <Zakim> Zakim has left #hcls2