15:09:28 <RRSAgent> RRSAgent has joined #hcls
15:09:28 <RRSAgent> logging to http://www.w3.org/2011/04/04-hcls-irc
15:10:53 <michael> michael has joined #hcls
15:11:29 <Zakim> +michael
15:13:53 <mscottm> scribenick: lena
15:14:12 <jun> http://purl.org/net/arrayexpress/ns
15:14:17 <mscottm> http://open-biomed.sourceforge.net/arrayexpress/owldoc/
15:14:29 <mscottm> http://purl.org/net/arrayexpress/ns
15:14:50 <Lena> rdf version of arrayexpress to geo
15:15:01 <Lena> (gene expression atlas)
15:16:23 <Lena> jun: not describe the experiments, about 100 experiments for a case study, transform sdrf file and send the processed data into rdf
15:16:41 <Lena> jun: create a picture of what probes were studied, which sample, which organism
15:17:03 <Lena> jun: provenance information like which experiment - for validating data
15:17:22 <Zakim> -EricP
15:17:24 <Lena> jun: opmv
15:17:48 <Lena> jun: reused terms from provenance use case
15:18:05 <Lena> jun: more interest in individual samples
15:20:56 <Lena> jun:provenance use case need to have samples'
15:21:21 <Lena> jun:  refactoring from helen parkinson example
15:23:38 <mscottm> Scott: We might want to take a look at how AO applies http://code.google.com/p/annotation-ontology/wiki/Homepage , possibly articulates with OPVM
15:34:34 <Lena> michael: many of the samples are technological replicates
15:34:47 <Lena> they go into the same bins of experimental factors
15:35:01 <mscottm> Scott: Could we use your arrayexpress schema to gain insight / compare to other versions such as GXA?
15:35:24 <mscottm> Jun: Well.. yes. But it might not be that interesting/applicable.
15:36:18 <mscottm> Scott: I suggested last time that we compare the RDF from the Provenance paper (ready), and compare it with the same data from GXA (Gene Expression Atlas) and NIF (Neuroscience Information Framework) versions.
15:36:25 <mscottm> Jun: Yes, that sounds good.
15:37:36 <Lena> michael: if two samples have the same experimental factors they are replicates 
15:42:23 <michael> replicate experiment: E-MIMR-13
15:43:59 <Lena> michael: we need to make sure we capture the exp factors
15:44:04 <Lena> and for each experiment
15:47:38 <Lena> Lena: users of the provenance must be able to add new experimental factor and indicate that they should be considered experimetal factors (like the others modelled by the ontology)
15:48:21 <michael> if two samples have the same experimental factor values(!) they are replicates
15:48:22 <Lena> I have to go... 
15:48:26 <Lena> bye...
15:48:33 <Zakim> - +1.832.386.aacc
15:49:34 <mscottm> https://docs.google.com/document/d/1A5-3tOsifPWPpETBKU-ZA9d7O7wK_nBzTFUBEe-0Bzo/edit?authkey=CK-y8Y8C&hl=en#
16:06:51 <jodi_> jodi_ has joined #HCLS
16:06:55 <Zakim> -[IPcaller]
16:06:57 <Zakim> - +1.216.368.aabb
16:06:59 <Zakim> - +20416aaaa
16:07:00 <Zakim> -michael
16:07:00 <Zakim> SW_HCLS(BioRDF)11:00AM has ended
16:07:02 <michael> michael has left #hcls
16:07:02 <Zakim> Attendees were michael, +20416aaaa, [IPcaller], EricP, +1.216.368.aabb, +1.832.386.aacc
16:07:08 <jodi> jodi has joined #HCLS
16:07:17 <jodi> jodi has left #HCLS
16:08:24 <jodi_> jodi_ has joined #HCLS
16:09:52 <mscottm> Scott: I was hoping to ask ericP about procedural matters and the w3c note. Maybe next time.
16:09:57 <jodi_> jodi_ has left #HCLS
16:11:03 <mscottm> Scott: One thing left to do is the RDF version of gene lists that corresponds to the example in the MAGE-TAB that Michael wrote.
16:11:23 <mscottm> Satya: What do you mean?
16:15:15 <mscottm> Scott: I mean using a list construction (RDF collection) to contain the gene list in such a way that users can query to get gene lists.
16:16:28 <mscottm> Satya: There are two ways to do that: one way would use a blank node which makes querying harder. The other way is to name the collection with something like 'import_list'. 
16:18:21 <mscottm> Scott: The second naming approach (i.e. avoiding blank nodes) sounds like the best approach. Would you like to write in that part of the document?
16:18:29 <mscottm> Satya: Yes.
16:18:37 <mscottm> Scott: Ok, I've added you.
16:21:44 <ericP> is there a principled way to name the nodes?
16:23:02 <mscottm> I think that the idea was just to coin an ad hoc name for the collection.
16:23:34 <ericP> oh, not the individuals in the collection?
16:23:52 <mscottm> no, at least that's not what I understood. 
16:24:10 <mscottm> While you're here :) can you tell us about any timing issues for the w3c note?
16:24:57 <ericP> while i like the idea of naming everything you can, i think you can't use either turtle's () syntax or rdfxml's parseType="Collection" if you name the collection
16:25:27 <ericP> a note can be published around 1 week after the IG decides to publish
16:25:54 <ericP> must be valid html and conform to the publication style and all that
16:26:07 <mscottm> It looks like we will have something to move into html in about a month.
16:26:31 <ericP> cool
16:26:54 <mscottm> It's optimistic but I was born that way.
16:27:52 <mscottm> Scott: I will follow up on trying to get our NFT data sets (Provenance paper sets) into GXA and NIF so that we can compare RDF.
16:28:29 <mscottm> Michael: I will probably get a chance to talk with Helen at a MAGE-TAB editor meeting at Stanford pretty soon.
16:28:45 <mscottm> Scott: Excellent. F2F is always good.
16:29:28 <mscottm> Scott: Will also contact Jim McCusker to see if he can supply his perspective.
16:30:30 <mscottm>   RRSAgent, please draft minutes
16:30:30 <RRSAgent> I have made the request to generate http://www.w3.org/2011/04/04-hcls-minutes.html mscottm
16:30:35 <mscottm>   RRSAgent, please make log world-visible
17:24:34 <Zakim> Zakim has left #hcls