HCLSIG/Meetings/2010-11-04 Conference Call

From W3C Wiki
Jump to: navigation, search

Conference Details

  • Date of Call: Thursday November 4, 2010
  • Time of Call: 11:00am Eastern Daylight Time (EDT), 15:00 British Summer Time (BST), 16:00 Central European Time (CET)
  • Dial-In #: +1.617.761.6200 (Cambridge, MA)

[NOTE earlier times for Europeans]

  • Dial-In #: +33.4.26.46.79.03 (France)
  • Dial-In #: +44.203.318.0479 (UK)
  • Participant Access Code: 4257 ("HCLS").
  • IRC Channel: irc.w3.org port 6665 channel #HCLS (see W3C IRC page for details, or see Web IRC), Quick Start: Click on mibbit for instant IRC access.
  • Duration: ~1h
  • Convener: Scott
  • Scribe: TBD

Agenda

Minutes

<mscottm> scribenick: Bob
Scott: Monday is BioRDF
... Kei has stepped down as chair due to time
... want to make use of momentum, progress in microarray data in both research and federation
... next Monday, will move toward consensus of RDF services
<matthias_samwald> i just joined
Scott: Pistoia has done federation and microarray
... CSHALS was talking re integrating experimental data in RDF from different perspective
... Have published at ISWC
... approach to both microarray and federation
... want to harmonize w details of experimental design, in sci discourse
... also align w. Pistoia project
... This all with an aim toward establishing consensus re information requirements
... BIoRDF supplies information re which genes selected
... refer to those w terms from an EBI ontology
... see What are the differentailly expressed genes, and with what confidence?
... Another aspect, to look at how to federate knowledge resources
... query from sparql endpoint, (currently-out-of-band) information
... there are 200 URIs, display of named graph should be able to look up the ontology
<mscottm> http://esw.w3.org/HCLSIG_BioRDF_Subgroup/Meetings/2010/10-25_Conference_Call
Scott: Terminology. Matthias has created special flavor of tags for mammogram use case.
... this is ongoing work.
... Eric, COI
Eric: Two things. NIH is interested in annotating web services
... wiki page is here:
<ericP> caBIG Semantic Workflows
Scott: Mark Wilkiinson(?) talking re importance of clinical data integration
... day-to-day basis in clinical environment
... COI came up in discussion, He's interested in moving COI forward
Eric: (discussing the wiki page)
<mscottm> http://www.w3.org/2007/Talks/0511-egp-spdl/
Eric: w/o xslt, fairly efficient and minimal curation process to ask sparql and get answers
Scott: Mark W. has sadie(?) system
<mscottm> SADI
Scott: SADI will find predicates that do not match perfectly
... SADI can fill in the gaps in sparql queries
Michel: We have a shared client, other code to do service invocation
... it does some choreography as well
... we can walk thru what drug-like molecule is
<michel> http://www.slideshare.net/micheldumontier/2010-cascon-towards-a-integrated-network-of-data-and-services-for-the-life-sciences
Scott: Ex is creatinine level; data can tell you rising for falling by reasoning
Eric: Terse annotations and graph-based reasoning synthesis will produce very interesting tool
... Should have discussions w Mark and Michel re ease of maintanance, expressability
Scott: LODD
... LODD there is an article, Eric gave talk at ACS, brougth semweb to chemistry
... Matthias is organizing the paper
... want to hear back from Anja re scripts
... generally, desire to come up w practice and ways to connect to experimental data
... will feed requirements from BioRDF to LODD for federation
... TMO
Michel: TMO is in the home stretch; Working on representation of patient data: Eric
... working w Indivo patient records
... large portion of attention on patient data, specialized predicates, how do they map?
... associate w patient record, IDs to other datasets
... focus now is on the paper
... Matthias has been looking at Big OWLim to check consistency
... Big OWLim can do it, a little hard
... will be better platform than our current platform
Scott: Feature request to Big OWL?
Michel: Answer is unknown yet...
Scott: Got an example from Indivo patient records: thanks Eric!
Eric: Unit testing was to handle dev of the patient data ontology
... we don't have actual patient data from Indivo, so we are giving them ours
... main goal is that they include grddl
... promote TMO to Indivo thru their grddle
Scott: Might Indivo reach a turning point?
Eric: Principal architect is interested now
... also Children's Hospital, Harvard will query i2b2 stores, we would like to integrate there
<mscottm> http://www.swat4ls.org/2010/index.php
Eric: integrate i2b2 w. Indivo (cool!)
... Possibility that SGI code will use it, TBL's group, Helen maybe
Scott: Still interested in working with Eric on mini-tutorial Dec 10
... take existing data, show how to configure
... we should show small, less ambitious examples
... NCI John Madden has job going to suggest arch. for federation of triple-stores in tissue banking
... also triple-stores: There is IMI grant from Lieden, 20M euro, lots of partners, firmly based on nano-publication provenance
... will write-in W3C into this proposal:
... trying to get co-ordination w. Pistoia.
... Barend Mons wants to put together a consortium: HCLS, Concept-Web, Pistoia, others!
... People can be puzzled by COI, need clinical reports in patient records (have I got this right?)
... use of clinical reports does not belong only to COI. Activities will coalesce.
... Thanks everybody!
<mscottm> Thanks Bob for scribing!