HCLSIG/Meetings/2010-09-02 Conference Call

From W3C Wiki
Jump to: navigation, search

Conference Details

  • Date of Call: Thursday September 2, 2010
  • Time of Call: 11:00am Eastern Daylight Time (EDT), 16:00 British Summer Time (BST), 17:00 Central European Time (CET)
  • Dial-In #: +1.617.761.6200 (Cambridge, MA)

[NOTE new numbers for European dial-in]

  • 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

Kei (BioRDF update): We are working on a microarray use case paper incorporating gene lists in RDF to be submitted to a provenance workshop associated with ISWC in Shanghai
... Gene lists are currently only unofficially available, not available at NCBI's GEO or EBI's ArrayExpress.
... We have created an approach to representing gene lists in RDF as they are associated with a microarray study
... We have been in touch with microarray standards such as MIAME and MGED (now called FGED, for Functional)
... The deadline for the paper is tomorrow
Scott: I hope that this important work will be taken up by the community and lead to omic data integration practices
Daniel Rubin (Radiologist/Terminology update): John Madden and I are working on representing information contained in mammogram reports from radiology and pathology using text markup ideas to create RDF and eventually represent reports in RDF that will allow us to use Semantic Web to detect discrepancies. I also have an image database that has Semantic Web access that could be eventually linked to the text markup
Daniel (introducing self): Radiologist from Stanford interest in SW approaches to images and clinical information
EricP (COI update): COI has been on vacation, looking at a performance paper for the COI and SWObjects (?). Also, look at handling patient records in COI, using TMO ontology
Scott: Yes, it would be good to make sure that we take advantage of overlap between task forces
Susie: TMO essentially aggregates ontologies and vocabularies but haven't used HL7 there yet
Scott: asks about Indivo for patient records
Eric, Susie: Indivo is broader than many commercial patient record systems and open source in XML, but we want to consume RDF
Susie: also want to map to CPR (Chime's)
EricP: Indivo also wants access to I2B2 data, so we'll all be sitting at the same table
Michel: We could take those XML elements and represent them in OWL
EricP: We can map to CPR and what ever ontologies we need, and use GRDDL/XSLT to map
Susie: Are there any efforts to align I2B2 and Indivo?
EricP: Nothing specific yet
<ericP> do mice have PatientEncounters?
Susie: There's no easy way to incorporate Next Gen or Mass spec data into I2B2. It would be interesting to see how to address this.
EricP: Maybe address this from TMO.
Susie (TMO update, see above): TMO has been on summer recess so the previous discussion covers it. We would like to look at how to represent patient records in TMO.
scribe: Also should mention that Michel's TMO presentation went well and is leading to a publication in JBS
<ericP> is cross-membership sufficient?
Michel: I would like to make sure that the use of ontologies used between COI and TMO are coordinated.
Susie: I agree.
Michel: We should try to ensure the coordination by having a meeting or something.
Joanne: I am concerned about siloization in task forces.
<ericP> ideally we can lean on indivo tools some for interface
Scott: We have been discussing 'normalization' across task forces in the form of a demo that makes use of artifacts from each task force (leading to coordination of vocabulary use)
<ericP> +1 to demos clarifying sharing opportunities
... This HCLS call is our opportunity to address these issues (and we've been doing so).
<jluciano> If we can't get it right, we are not setting a very good example
Susie (LODD update): LODD has been on recess but is indeed interested in those issues
<jluciano> I agree with what susie is saying now-need a more comprehensive strategy to coordinate among hclsig task groups
Susie: Scott and Eric and I have had weekly calls (somewhat less lately) to coordinate as well.
<ericP> mscottm: per TF coordination, how do we pick an ontology for representing e.g. diseases?
EricP: I think that these HCLS calls aren't frequent enough to satisfy the needs of the task forces
... It seems that LODD has the most experience with using LOD.
<ericP> mscottm: in BioRDF, we're addressing provenance for SPARQL endpoints
<ericP> ... these might be general enough to be a "best practice" and maybe include in the LODD Best Practice doc
<ericP> matthias_samwald1: ontology bashing should occur on the mailing list
<ericP> Susie: agreed, if we pick a disease ontology, we should talk about it on the mailing list as well as a 1st Thu call
<ericP> kei: this is a tough problem. Alzheimer's shows up in probably 30 ontologies
<ericP> ... could be a long debate
<ericP> ... we could come up with some objective [ontolgy evaluation] criteria
<ericP> Susie: if we've developed an ont within HCLS, i'd like to see that used across task forces
Susie: I think you make some good points. If TMO develops an ontology, it would be best for HCLS to use it.
<ericP> ... per outside ontologies, i appreciate it could be hard to maintain consistency across TFs
<ericP> ... perhaps a wiki page listing the HCLS ontologies
<ericP> jluciano: should we have a group of people responsibile for coordination?
<ericP> mscottm: not sure we need a coordination task force
<ericP> ... it's unfortunate when IDs used in one TF are not used in another
<ericP> i want to have a coordination task force. i think michel should have one as well, and we could have a coordination task force coordination task force.
General Consensus: create a wiki page to coordinate vocabulary choice (and other issues)
Scott: We will have to discuss the FDA CDER Data Standards Plan in another venue
Joanne: Please remind on the list