HCLSIG/Meetings/2009-04-02 Conference Call

From W3C Wiki
Jump to: navigation, search

Conference Details

  • Date of Call: Thursday April 2, 2009
  • 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)
  • Dial-In #: +33.4.89.06.34.99 (Nice, France)
  • Dial-In #: +44.117.370.6152 (Bristol, 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)
  • Duration: ~1h
  • Convener: Scott, Susie

Agenda

Minutes

  • Attendees: John, Jamie, Scott, Bosse, Kevin, Michel, EricP, Stefan, Jun, Egon, Oliver, Marco, Matthias, Susie
  • Apologies: Kei
  • <ericP> topic: task force updates
  • <ericP> topic: BioRDF [ScottM]
  • <ericP> mscottm: working on a BMC Bioinformatics paper (due monday)
  • <ericP> ... topics included ATAGS, Aida use of HCLS DERI KB
  • <ericP> ... (same interface used to browse JohnM's SKOS)
  • <ericP> ... discussion of merging jun's herbal DB with LODD datasets
  • <ericP> ... will load those into the KB
  • <ericP> jun: discussed traditional chinese datasets with Anja
  • <ericP> ... dataset comes from text mining, so proceeding cautiously
  • <ericP> ... re f2f: confirmations from John Breslin, Michel Dumontier
  • <ericP> jun: talked about demo of AIDA access to deri hcls kb
  • <mscottm> http://ws.adaptivedisclosure.org/search/
  • <mscottm> http://tarski.duhs.org:8080/openrdf-sesame
  • <mscottm> In Server Details: enter tarski link
  • <mscottm> click on Repositories
  • <mscottm> http://adaptivedisclosure.org
  • <ericP> michel: am working with Bio2RDF group to expand the dataset and make the data more meaningful
  • <ericP> ... tying to ontologies and unifying terms
  • <ericP> johnM: is @@1 using quads?
  • <ericP> mscottm: when you click on instances, you see the named graphs
  • <ericP> ... many named graphs, and you can't guess what's in them
  • <ericP> ... in general, federation requires named descriptions of repositories *and* named graphs
  • <ericP> topic: LODD
  • <ericP> Susie: working on LODD paper for WWW
  • <ericP> ... thinking about what we not to do next
  • <ericP> ... goal: submit a paper to the iTriplification challenge (due May 30)
  • <ericP> ... working on connecting to other datasets, e.g. dbpedia's large corpus of company data
  • <ericP> ... looking at traditional chinese datasets
  • <ericP> ... looing at pharma ontologies role questions to pick the 3 most interesting
  • <ericP> ... will help us choose a couple more datasets to add to the cloud
  • <ericP> ... (at a lower priorty) looking for more consistancy in mappings so using pharma onts
  • <ericP> ... want a hybrid interface for not only browsing, but querying
  • mscottm provides details that didn't flow during the task force update: The resources that we are querying in BioRDF paper are: the KBs, SenseLab, aTags, Faviki, IUPHAR
  • <ericP> ... .. adding template queries e.g. what do you know about this gene?
  • <ericP> ... LODD will not meet at f2f as many europeans won't make it
  • <ericP> ... pharma ont will meet and will be working on onts useful to LODD
  • <ericP> topic: pharma ont
  • <ericP> Susie: goal: create a template ontology for plugging in existing/upcoming onts
  • <ericP> ... so we can generate queries which span covered areas
  • <ericP> ... working on what questions folks need to ask and what roles they imply
  • <ericP> ... expect to finalize the roles and their questions in the next call (35 mins)
  • <ericP> ... expect ~ 40 entities in ontology
  • <ericP> ... given entities and use cases, we'll build out the ontology in that area and build a tool to use it
  • <ericP> ... submitting abstact biomedical ontologies and @@2 at ISMB
  • <ericP> ... encouraing companies to share their internal ontologies
  • <ericP> ... these (e.g. lilly's) are developed in isolation so looking at adopting e.g. BFO practices
  • <ericP> topic: COI
  • <ericP> HelenC: ericP produced SWtransformer and Holger and I worked on an interface
  • <ericP> ... we submitted to AMIA conference
  • <ericP> ... will look at a more tech opportunity e.g. WWW or ISWC
  • <ericP> ... will be at the f2f to examine query federation
  • <ericP> ... canadian research consulate (member of W3C?) has a large group working in semweb
  • <ericP> ... talking with them

<ericP> ... next week will discuss ops for new collaboration <ericP> ... looking for use cases and data of clinical relevence

  • ericP tx HelenC!
  • <ericP> topic: Terminology
  • <ericP> johnM: we have a couple repos on the server working with AIDA
  • <ericP> ... want a 2nd ont with RDFS turned on so we can do on-the-fly terminologies
  • <ericP> ... last time we had the repo up, but AIDA was having probs accessing repo. will have a webcon to debug
  • <ericP> ... we're using the native sesame RDFS inferencer
  • <ericP> ... trying to use OWLIM, but it's difficult to get it to talk to sesame
  • <ericP> ... federation is still on my plate
  • <ericP> ... we got Simon Schenk and his lightweight federation gear
  • <ericP> ... we have all these terminologies, but we need data
  • <ericP> ... have been using scott's text data
  • <ericP> ... but want SNOMED-encoded data
  • <ericP> ... just spoke to a collegue at Duke who has to large repos set of RDBs
  • <ericP> ... 1: Duke tissue bank
  • <ericP> ... 2: anotomoical? DB
  • <ericP> ... both are up on web-accessible RDBs
  • <ericP> ... will ask him to install d2r server
  • <ericP> ... style of annotating medical records is flat (no S P O triples)
  • <ericP> ... basically same as the needs driving the pharma ont
  • <ericP> ... haven't worked with Chime's [CPR] patient ontology yet. will look at it and associated use cases
  • <ericP> ... two deadlines ~ Sept:
  • <ericP> ... .. 1 will be platform speakers at AMIA/APIII
  • <ericP> ... .. .. will talk about semantic extraction (using GRDDL)
  • <ericP> ... .. .. can use a the patient ontology
  • <ericP> ... hope to wow audience with a soup-to-nuts semweb application
  • <ericP> topic: f2f
  • <ericP> mscottm: anybody here who isn't going to f2f?
  • <ericP> -> http://esw.w3.org/topic/HCLSIG/Meetings/2009-04-30_F2F HCLS F2F
  • <ericP> Susie: a bunch of folks at onto tf: Elgar Pichler, Bosse, and christi dialing in
  • <ericP> mscottm: BioRDF plans to spend a couple hours to geek with johnM about HC terminolgies
  • <ericP> [Susie reviews agenda]
  • <ericP> R&D open 'till 10pm
  • <ericP> Susie: please add your names to the attendance list, and inidicate if you will be calling in
  • <mscottm> http://www.slideshare.net/micheldumontier/accurate-biochemical-knowledge-starting-with-precise-structurebased-criteria-f...
  • <ericP> -> http://esw.w3.org/topic/HCLSIG/Meetings/2009-04-30_F2F#attend attendance list
  • <ericP> topic: michel's preso to NCBO
  • <ericP> mscottm: this is a structure-based approach to chemical identificationh
  • <jun> i was there. it was a really good talk:)
  • <ericP> topic: misc
  • <ericP> Susie: sometimes we drill into a particular task during the general HCLS calls (thursdays) or invite presenters
  • <ericP> ... if you have suggestions or requests, please mail the chairs list team-hcls-chairs so we can try to please everyone
  • <ericP> marco: Concept Web Alliance is working on a large hcls use case
  • <ericP> mscottm: a bit in parallel with hcls
  • <ericP> [susie and marco will represent hcls use cases]
  • <ericP> mscottm: is similar in goals to this group
  • <mscottm> CWA = Concept Web Alliance = HCLS IG in another flavor
  • <ericP> Susie: could be synergistic with hcls
  • <ericP> ... if they get a wide group thinking about what they want from an eScience environment, hcls can help fill in missing pieces