HCLSIG/LODD/Meetings/2009-02-18 Conference Call
- Date of Call: Wednesday February 18, 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
- Update on the paper
- Triplification challenge
- What's next?
- Follow up on the issues identified in the outlook section (browser performance and functionality, identifier ambiguity, platform for reasoning).
- Adding new data sources as proposed (SWAN hypothesis, epidemiology, companies, Medpedia).
- Address the need to capture user input
- Explore how applications like LLD (LarKC) can add value into the Linked Data cloud.
- Links to similar compounds
- Improve quality of the data sets
Attendees: Oktie, Jun, Chris, Don, Kei, Anja, Susie
Apologies: Matthias, Bosse
<kei> susie: 1. update on paper
<kei> susie 2. what's next
<kei> susie: paper was submitted
<kei> susie: thank to everyone's hard work
<kei> susie: when will we hear
<kei> chris: 30+ submissions
<kei> chris: notification date -- March 1st
<kei> susie: need to get permission to be a co-author on the paper -- it was approved
<kei> chris: matthias sent email about an interface to the lodd datasets?
<kei> chris: crawl the datasets and put them in triplestore
<kei> chris: marble too slow
<AnjaJentzsch> Example query: http://visinav.deri.org/hcls/list?keyword=varenicline "
<kei> susie: share with scientists -- asking them to test and validate and for feedback
<kei> chris: a lot of info displayed for drugs
<kei> chris: need more provenance info. but a nice interface
<kei> susie: people should play with the interface and give feedback ...
<kei> don: where is the data from?
<kei> Anja: sparql endpoints
<kei> susie: main home for the data sources should be accessed more obviously
<kei> Susie: can Anja take care of this?
<kei> Anja: Andreas became aware of LODD data sets through Richard.
<kei> susie: next step
<kei> susie: one next step is to make sure data connections are good
<kei> susie: google doc was written about this
<kei> susie: ask people in Lilly to address the data connection issue
<kei> susie: Don knows anything about how to connect generic drugs vs. specific drugs
<kei> Don: can look into it (e.g., pdb contains information about binding sites)
<kei> susie: give Don access to google doc
<kei> chris: need someone who knows more about the data and how to use the data to guide us
<kei> chris: devise rules for linking data sets
<kei> chris: some example rules based on similar names, similar properties, etc
<kei> susie: Lilly people are excited about using this approach to find out more about compounds (developed internally and externally)
<kei> susie: how rules help to identify similarity/difference between internally developed compounds and externally developed ones?
<kei> susie: need to work with domain experts in developing the appropriate rules ...
<kei> anja: bio2rdf focuses on chemical compound related data -- could be a good resource
<kei> chris: link to bio2rdf instances -- so no need to store such instances locally
<kei> susie: chemical compounds exist in public repositories or lab-specific databases
<kei> susie: tighten up the links ...
<kei> susie: better develop the rules ...
<Susie> Triplify challenge: http://www.i-semantics.tugraz.at/triplification_challenge
<kei> oktie: semantic conference in austria -- submissions about using linked data
<kei> oktie: our application is a nice demonstration of this challenge
<kei> oktie: deadline for submission to the conference is May 30
<kei> susie: is this a conference paper?
<kei> oktie: it's only a 2-page report on challenge.
<kei> chris: it's appropriate
<kei> susie: also think it's appropriate
<kei> oktie: we may need to think more about ontologies
<kei> oktie: coming from data management background, would be nice if someone with strong ontology background can participate ...
<kei> susie: how ontology helps in aggregation of linked data?
<kei> chris: put all data out as first step, next step people starts exploring data, ontology mapping and the like ...
<kei> chris: we are near the end of the 1st phase
<kei> chris: might use ontology to help browsing/mapping
<kei> oktie: two clouds: data cloud and class-level interlinking cloud
<kei> susie: can more information about this new cloud be available?
<kei> oktie: schema of data
<kei> chris: can find high level information but not more detailed info...
<kei> susie: may make sense to take advantage of ontologies ...
<kei> chris: don't need to agree on global ontology
<kei> chris: simple/light-weight one may make more sense ...
<kei> susie: pharma-ontology work may help ..
<kei> susie: disease and patient centric not drug related in this case ...
<kei> susie: lodd might be a potential use case for pharma ontology
<kei> susie: good to see deliverables in the pharm ontology group
<kei> susie: update the web site to mention the challenge for the conference
<kei> susie: other things to discuss
<kei> susie: outlook section -- identify ambuguities, etc ...
<kei> susie: build up a robust platform for reasoning
<kei> susie: capture user input -- adding more content to the cloud.
<kei> susie: exploring datasets to add to the cloud ...
<kei> susie: hypothesis data
<kei> susie: ask elizabeth wu and tim clark
<kei> susie: adding more data regarding epidemiology ...
<kei> susie: more wiki related resource just started ...
<kei> susie: data source contributed by experts ...
<kei> Susie: ask Don to present in 2 weeks
<kei> Susie: ask Jun and Kei to present in 2 weeks
<Susie> Jun: presented on void during BioRDF call
<Susie> Jun: Looking for community feedback
<Susie> Jun: Not many tools yet
<Susie> Jun: Very early to be using void
<Susie> Susie: I'll be giving a brief presentation on LODD to CSHALS next week