HCLSIG/LODD/Meetings/2009-02-04 Conference Call

From W3C Wiki
< HCLSIG‎ | LODD‎ | Meetings

Conference Details

  • Date of Call: Wednesday February 4, 2009
  • Time of Call: 11:00am Eastern Daylight Time (EDT), 16:00 Greenwich Mean Time (GMT), 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: Susie
  • Scribe: Anja

Agenda

  • Data set recap - Anja
  • Use case discussion - Susie & Bosse
  • Plans for writing the paper - All
  • AOB

Minutes

Attendees: Egon, Elizabeth, Don, Kei, Bosse, Chris, Anja, Oktie

<Susie> Anja gives a review of the work she's done

<Susie> Anja: 3 data sets RDFized

<Susie> Anja: DrugBank on generic drugs

<Susie> Anja: DailyMed on brand drugs

<Susie> Anja: Diseasome on diseases and associated gene

<Susie> Anja: All data sets can be linked

<Susie> Anja: Links to Bio2RDF: pubmed, KEGG, etc.

<Susie> Anja: Have linked to dbpedia

<Susie> Anja: And can link to LinkedCT.gov

<Susie> Anja: Need connections to LinkedCT

<Susie> Chris: Drugbank is mainly interlinking hub as can get to diseasome, DailyMed, Bio2RDF

<Susie> Anja: Now need to focus on the path through the cloud

<Susie> Anja: LinkedCT is currently just linked to dbpedia

<Susie> Anja: Several links between data sets

<Susie> Anja: Using different identifiers

<Susie> Anja: If see error please let Anja know

<Susie> Susie: Susie & Bosse in IT aren't experts in terms of working with the data

<Susie> Susie: Will guide as best as we can, but will want to share work with scientists

<Susie> Anja: Can view in HTML at the Free University of Berlin

<Susie> Anja: But it's not ideal

<AnjaJentzsch> Marbles link for Alzheimer's disease: http://beckr.org/marbles?uri=http://www4.wiwiss.fu-berlin.de/diseasome/resource/diseases/74

<Susie> Anja: Are using Marbles for the mashups

<Susie> Chris: But can be slow because there is so much data

<Susie> Chris: Need a bigger machine for better performance

<Susie> Chris: Currently, running on a machine in an Amazon could that isn't powerful enough

<Susie> Susie: Has anyone come across http://squin.sourceforge.net/

<Susie> Susie: Sometimes hard to know how to define a use case for browsing rather than querying

<Susie> Chris: Navigation allows you to discover things that you aren't aware of

<AnjaJentzsch> Varenicline (drug) on Marbles: http://beckr.org/marbles?uri=http://www4.wiwiss.fu-berlin.de/drugbank/resource/drugs/DB01273

<Susie> Chris: To get the paper published we need to have a good use case and motivation

<Susie> Mathhias: Could identify shortcomings

<Susie> Chris: Can't focus on known problems

<Susie> Chris: We didn't cache the data so it's not suprising that it's slow

<Susie> Bosse: challenge in integrating data exists with or without linked data

<Susie> Bosse: Team with different expertise can navigate the data and find things relevant to their domains

<Susie> Bosse: This isn't possible with other technologies

<Susie> Bosse: Need vision, and then we should be able to demonstrate it

<Susie> Egon: Has anyone looked at derived data?

<Susie> Susie: Potentially start a use case from the idea of forming a collaboration

<Susie> Kei: Could be interesting

<Susie> Bosse: Normally not interested in business oriented things

<Susie> Bosse: If designing a trial, you want to see what others are doing, and you can learn from it

<egonw> Susie, don't want to knock you down at all, but I think pharma will not share compound activities they find really interesting...

<Susie> Chris: Could include 2 use cases

<Susie> Bosse: Can work on a scenario and see if it's possible on Thursday or Friday

<egonw> I think I could bring in a use case from data modeling, where aggregating data is typically a problem

<Susie> Chris: More details is better

<Susie> Bosse: Agree on vision of crossing domains

<Susie> Chris: Tuesday is very final deadline

<Susie> Bosse: Finishes pharma specific motivation by lunchtime on Saturday

<Susie> Chris: I'll add generic Linked Data, and reference to Bio2RDF, LinkedCT, etc by Sunday evening

<Susie> Susie: I'll aim to complete collaboration use case by end of Thursday, but it may end up being Friday.

<Susie> Bosse: I'll aim to finish use case section by Sunday

<Susie> Anja: I'll finish data set section by Saturday

<Susie> Susie: I'll draft the outlook section by Sunday evening

<Susie> Susie: If anyone has thoughts about the outlook section please add them to http://esw.w3.org/topic/HCLSIG/LODD/Business

<Susie> Susie: Lets write the paper in Word

<Susie> Susie: I'll set up a google doc

<Susie> http://beckr.org/marbles?uri=http://www4.wiwiss.fu-berlin.de/drugbank/resource/drugs/DB01273

<Susie> http://beckr.org/marbles?lang=en&uri=http://www4.wiwiss.fu-berlin.de/diseasome/resource/diseases/74

<Susie> Kei: Nice to have a bookmark as starting point to the Linked Data

<egonw> Egon: a greasemonkey script could help here...

<egonw> Egon: which would add a link to LinkedData from wikipedia pages (using DBPedia)

<egonw> Susie: egon.willighagen@gmail.com

<egonw> Egon: it's indeed a short time frame... but might still write up thought, which I would email, not add to the GDoc

<AnjaJentzsch> we have the d2r servers for each dataset as starting points

<AnjaJentzsch> http://www4.wiwiss.fu-berlin.de/diseasome/

<AnjaJentzsch> http://www4.wiwiss.fu-berlin.de/drugbank/

<AnjaJentzsch> http://www4.wiwiss.fu-berlin.de/dailymed/

<Susie> Oktie: happy to share links with Anja

<Susie> Oktie: happy to contribute worked on LinkedCT to LODD

<Susie> Susie: Wonderful!

<Susie> Oktie to be a co-author

<Susie> Oktie to add paragraph on LinkedCT by Sunday