See also: IRC log
<matthiassamwald> (I am away for 1 minute...)
<dbooth> Scribe: ericP
<matthiassamwald> (back again)
Susie: DSE in google docs
... we have STDM discussion in the note
<AdrianP> DSE note: http://docs.google.com/View?docid=dthvctw_5fwwwprc3
eneumann: note is a discussion of
issues around clinical trial data
... this is a broad area with lots of submission
standards
... so we're focusing on the STDM
... asking "what features would SemWeb bring to it?"
... intended to be a walkthrough for this community
... current exchange is a Sas (SP?) table
... there is an XML structure for the STDM
<alanr_> ack ??P1
eneumann: issues arrise from that; discussed with Bo and Kersten
alanr: hoped to capture that in
the abstract
... yellow indicates questions and further development
... expect 3 weeks to pub if we get Bo, eneumann and kersten
time
eneumann: i would like Bo and Kersten to record their issues with the current STDM in the abstract
AdrianP: i think we lost the
point that the original STDM has potential to meet our
needs
... we lost the point that the data is useful in the XML
STDM
alanr: wanted to clarify interoperability - interop between what?
AdrianP: it's hidden there
... [Biomarkers and clinical observations]
alanr: people doing the study are
generating the STDM stuff
... they submit this to the FDA, but our issue is the re-use
within the company?
AdrianP: yes
Susie: Lilly asserts that the
STDM is a very high-level model
... Lilly will use different terms for standing BP than AZ
will
... we outsource the trials to other companies (Quintiles),
stipulating the info and the coding (e.g. Lilly's STDM
terms)
... Quintiles, FDA and Lilly all hate this diversity of
terms
... Lilly has presented to CDISK that this be nailed down
<eneumann> ?+
<eneumann> +?
<Zakim> alanr_, you wanted to ask whether current abstract gets it right
Susie: considering using a wiki to unify these terms
kersten: spot on. exactly why i'm
working on this at AZ
... STDM has only the fragments of what we need
<eneumann> ss/CDISK/CDISC
kersten: will take up issues in Lilly presentation
Susie: our expert is not content with using lists to solve this problem
alanr: STDM code lists?
kersten: don't have ways to select from of STDM codes lists
<alanr_> susie: Any way to get someone from quintile(sp?) to come and complain to us?
kersten: also need model extensibility for e.g. room temp
<alanr_> specifics of how the different pharmas differ would be very helpful
<eneumann> regarding SDTM codes, a common complaint is that it has too many places for entering your own uncontrolled code
kersten: STDM provides ways to code the data but CDISC doesn't [help you chose them?]
<alanr_> SDTM has no way to constrain saying a statement of blood pressure should at least have x,y,z
<alanr_> ACTION: Kersten and Bo to clarify what is missing about interoperability in abstract. Expand section on modeling issues [recorded in http://www.w3.org/2008/01/28-BioRDF-minutes.html#action01]
i guess that means that STDM allows you to record many more parameters than you want CDISC to to specify as "standard"
eneumann: motivating probs came from XML encoding
<alanr_> Good - Interoperability as SDTM across CROs, FDA, departments in a pharma
<alanr_> Missing is that semantics and modeling not clear enough to do so now
eneumann: Jim (Merck) wanted STDMs to be extensible when necessary
<alanr_> ericn: Another purpose of extensibility is to do analytics on the SDTM, which becomes more interesting when you can augment what's there with additional information easily
Susie: challenge is when to standardize and when to let innovation flourish
alanr: can we get Quintiles explain what they need to do for the different pharma clients?
Susie: can get the Lilly person to do that and he can do it as well
<alanr_> ACTION: Susie to invite Lilly person to discuss problems with SDTM [recorded in http://www.w3.org/2008/01/28-BioRDF-minutes.html#action02]
<eneumann> Jim McGurk has left Merck and is now at Daiichi-Sankyo
kersten: there is an ODM model for STDM in XML, but still basically tabulated data
matthiassamwald: hasn't changed for two weeks
<scribe> ACTION: matthiassamwald to have note ready for pub 4 Feb [recorded in http://www.w3.org/2008/01/28-BioRDF-minutes.html#action03]
Susie: can you have the KB doc ready as well?
mscottm: folks want different
things from the KB note
... have to pick a target audience. do they know RDF? SPARQL?
OWL?
... would like folks on the list to sanity check the data
model
... have some reservations, but think we've neared what we can
expect to publish
... section of need for integration seemed compelling to me
<AdrianP> I think references to the main RDF, OWL specs are enough for readers of the note
mscottm: would like Susie, ericP
to review this week and then ask the list for review
... i have a request to double or triple the size of the
abstract
<AdrianP> have to leave soon, too
<eneumann> I have to run-- bye!
<scribe> ACTION: ericP to be done with KB doc EOB Thursday [recorded in http://www.w3.org/2008/01/28-BioRDF-minutes.html#action04]
Susie: maybe pub SenseLab and KB notes out 2nd week of Feb
<Zakim> ericP, you wanted to clarify features missing in XML STDM
Susie, extended workshop submission deadline by a week
scribe: haven't succeeded in
getting final word from Greg Tucker Kellogg
... need to give him a firm deadline
<Susie> http://ccnt.zju.edu.cn/hcls2008/
ADJOURNED
This is scribe.perl Revision: 1.133 of Date: 2008/01/18 18:48:51 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: s/CDISK/CDISC/ Succeeded: s/SSDTM/SDTM/ Found Scribe: ericP Inferring ScribeNick: ericP WARNING: No "Present: ... " found! Possibly Present: AdrianP Bo Eric_Neumann Kerstin_Forsberg P1 P24 P3 P32 P5 P9 alanr alanr_ dbooth eneumann ericP ericn inserted kersten matthiassamwald michel mscottm susie You can indicate people for the Present list like this: <dbooth> Present: dbooth jonathan mary <dbooth> Present+ amy Got date from IRC log name: 28 Jan 2008 Guessing minutes URL: http://www.w3.org/2008/01/28-BioRDF-minutes.html People with action items: bo ericp kersten matthiassamwald susie[End of scribe.perl diagnostic output]