HCLSIG/PharmaOntology/Meetings/2009-09-17 Conference Call

From W3C Wiki
Jump to: navigation, search

Conference Details

  • Date of Call: Thursday September 17 2009
  • Time of Call: 12:00pm - 1pm ET
  • Dial-In #: +1.617.761.6200 (Cambridge, MA)
  • Dial-In #: +33.4.89.06.34.99 (Nice, France)
  • Dial-In #: +44.117.370.6152 (Bristol, UK)
  • Participant Access Code: 42572 ("HCLS2").
  • IRC Channel: irc.w3.org port 6665 channel #HCLS2 (see W3C IRC page for details, or see Web IRC)
  • Mibbit instructions: go to http://www.mibbit.com/chat and click the server link. Enter irc.w3.org:6665 into that box, enter a nickname, and enter #HCLS2 for the channel
  • Duration: 1h
  • Convener: Susie

Agenda

  • Progress on time line - Susie
  • Select the use case to implement - all
  • Select use cases for grounding the ontology - all
  • Review of ontology - Colin
  • F2F meeting - Susie

Minutes

Attendees: Joanne, Chris, Matthias, Michel, Bosse, Christi, Julia, Trish, Anja, Chime, Tim

Apologies: Colin

<Susie> http://esw.w3.org/topic/HCLSIG/PharmaOntology

<michel> susie: roadmap to see what needs to get done by when

<michel> susie: this week - get the ontology in OWL - on wiki

<michel> susie: need to select use case

<michel> susie: by next week, consider relations

<michel> susie: and the data!

<Joanne> in the timeline, we seem to be missing the item "a miracle occurs"  :-)

<michel> susie: 2 weeks out implement ontology, extensions

<michel> susie: data in RDF

<michel> susie: 1 month TMO grounded on secondary use cases, what exactly will be shown

<michel> susie: 3 week of Oct, have first version of app complete with first draft of paper

<michel> susie: complete, realistic?

<Joanne> Michel: It depends on the selection of data source - if we are not choosing a new data source, we should be OK

<michel> susie: perhaps choose the data for the use cases

<michel> susie: use cases given the data

<michel> joanne: do we have data for the use cases?

<michel> susie: we'll have to assess as we go over them

<michel> susie: LODD could provide data

<AnjaJentzsch> LODD Top Questions http://www4.wiwiss.fu-berlin.de/lodd/topquestions/

<michel> susie: dbpedia, Freebase,

<michel> susie: consider other tools in the development of our application

<AnjaJentzsch> DBpedia Faceted Browser http://dbpedia.neofonie.de/browse/

<michel> susie: what's available in data, what tools are out there,

<michel> joanne: pharmgkb?

<AnjaJentzsch> Freebase Faceted Browser (Parallax) http://www.freebase.com/labs/parallax/

<michel> michel: our previous work RDFized pharmgkb data, so it is available

<michel> anja: links to other OBO ontologies

<michel> topic: use cases

<Susie> http://esw.w3.org/topic/HCLSIG/PharmaOntology/UseCases

<michel> Alzheimer, Chemogenomics, integrative informatics, animal models, pharmacogenomics

<michel> no links to targets/biomarkers

<michel> susie: incorporate classic TM segmenting on biomarkers, but should be broader - with economic, comparative effectiveness

<michel> tim: really about de-siloizing data

<michel> tim: alzheimer is promising as a use case, but not as currently stated

<michel> tim: comment on use cases - use case should indicate the activities of the scientist

<michel> joanne: current entry not enough

<michel> susie: would be good to have better Alzheimer's use case to demonstrate TM aspects

<michel> tim: key area; in pharma, chemogenomics on proprietary molecules, but unlikely to find public data

<michel> christi: if the ontology is available, it can be used internally

<michel> chris: what's important to pharma? must try to represent several aspects

<michel> susie: broad institute - connectivity map, gene expression profiles with compounds

<michel> susie: are expecting pharma to use ontology internally

<michel> would support interopt goals

<michel> http://www.ebi.ac.uk/chembl/

<michel> medicinal, clinical, launched therapeutics data

<Joanne> we can use Peter Tonelatto's (Laboratory for Personalized Medicine at Harvard) to generate artifical data sets of patient records with whatever demographic distribution we want.

<michel> tim: compounds, hit selection, pharmacogenomics/dynamics, human data

<michel> tim: can we thread through the various use cases with an example

<Joanne> he may have a set already we can use ... in the spring he did an experiment with the anticoagulant warfarin.

<michel> susie: exactly what question will be answered

<michel> tim: want people to ask questions and get useful answers

<michel> tim: make use of previous data

<michel> susie: earlier work was done to address what questions do people ask when in certain roles

<michel> http://esw.w3.org/topic/HCLSIG/PharmaOntology/Roles

<michel> susie: should we have a separate call to go through the questions

<michel> susie: relevant existing data - clinicaltrials.gov data, dailymed, drugbank

<michel> susie: traditional chinese dataset,

<michel> susie: pharmgkb

<michel> susie: HCLS kb

<michel> tim: report experiments and experimental results in a general way

<michel> susie: ontology is being used to annotate the data in a consistent manner

<michel> tim: reuse all prior data, most preclinical, with negative results, try to figure out what that happened, develop better screens. problem was that nobody knew where the data was

<michel> susie: some integration with SWAN?

<michel> tim: possible to work on something with interested parties

<michel> susie: schedule a longer call to talk about use cases

<Susie> http://esw.w3.org/topic/HCLSIG/PharmaOntology/Ontology

<Joanne> thanks Michel for scribing!