IRC log of hcls2 on 2011-10-06
Timestamps are in UTC.
- 16:03:25 [RRSAgent]
- RRSAgent has joined #hcls2
- 16:03:25 [RRSAgent]
- logging to http://www.w3.org/2011/10/06-hcls2-irc
- 16:03:33 [mscottm]
- Zakim, who is here?
- 16:03:41 [mscottm]
- Zakim, this is tmo
- 16:03:41 [bobP]
- bobP has joined #hcls2
- 16:04:03 [adrien]
- adrien has joined #HCLS2
- 16:04:04 [Zakim]
- sorry, mscottm, I don't know what conference this is
- 16:04:14 [Zakim]
- ok, mscottm; that matches SW_HCLS(TMO)11:00AM
- 16:04:20 [Zakim]
- On IRC I see bobP, RRSAgent, mscottm, matthias_samwald, ericP, Lena, Zakim, michel2
- 16:04:51 [Zakim]
- +??P55
- 16:05:03 [Zakim]
- +Bob_Powers
- 16:06:24 [Zakim]
- + +33.3.83.59.aabb
- 16:07:20 [bobP]
- scribenick bobP
- 16:07:28 [Zakim]
- +Tony
- 16:07:33 [BobF]
- BobF has joined #hcls2
- 16:08:09 [bobP]
- Michel: Look at use case template, think of task to perform
- 16:08:48 [bobP]
- ... this group can champion what we want to do with the KB
- 16:10:10 [bobP]
- Michel: Agenda: Matthias, develop use cases, and research
- 16:10:34 [bobP]
- Matthias: Thanks everybody for great collaboration
- 16:10:41 [adrien]
- thank you Matthias for submitting it!
- 16:10:53 [bobP]
- ... one of difficulties was that we did two things at once
- 16:11:25 [bobP]
- ... report on how semantics implemented, also design choices etc on the other hand
- 16:11:46 [Zakim]
- +??P7
- 16:12:01 [bobP]
- ... another paper will be research article in biomed semantics
- 16:12:19 [Zakim]
- +??P26
- 16:12:34 [iker]
- iker has joined #hcls2
- 16:12:53 [bobP]
- ... Special Report short paper, turned out to be longer, so wrote a proper review article
- 16:13:21 [bobP]
- ... produced a very interesting and useful piece of scientific literature
- 16:13:28 [bobP]
- ... can use as a basis to communicate
- 16:13:50 [bobP]
- Michel: Hit the level that we wanted; broad scope of utility
- 16:14:14 [bobP]
- ... readers will get the gist. Tech platform for what we are doing
- 16:15:03 [bobP]
- BobF: Thank you very much for driving this manuscript
- 16:15:51 [bobP]
- Matthias: Now that this is done, other paper should be easier
- 16:16:35 [bobP]
- ... can really focus on tech. Pgx in OWL will be very interesting
- 16:17:06 [bobP]
- ... pgx data and related data, at the end will be very interesting
- 16:17:41 [bobP]
- Michel: Would really like to see a list of design patterns for expressing data
- 16:18:19 [bobP]
- ... in line w. new interest group. More on the vocab side
- 16:18:37 [bobP]
- ... not much yet from w3c about representational patterns
- 16:19:08 [bobP]
- ... how to represent snp data, relation to pgx. Patterns, should be possible to marshal
- 16:19:34 [bobP]
- ... going to try to push this, beyond just strict RDF, but a bit more (disciplined)
- 16:20:04 [bobP]
- BobF: Absolutely.
- 16:20:17 [mscottm]
- 1+
- 16:20:20 [bobP]
- +1
- 16:21:12 [bobP]
- Michel: bio2rdf, need patterns for representation, not yet well developed
- 16:21:37 [bobP]
- ... we have always used obo ontologies in my research
- 16:22:07 [bobP]
- ... have a whole list of design patterns, then becomes like a recipe
- 16:22:19 [bobP]
- ... go the W3C and get the pattern.
- 16:23:43 [bobP]
- Michel: Next step. Use case template, would like to...
- 16:23:59 [michel2]
- https://docs.google.com/document/d/1I9xyVKhO9wG7My2fWu9S-wMwWZxG6PkRc98kOxUqH-o/edit?hl=en_US
- 16:24:46 [bobP]
- BobF: dbSNP last correspondence, mentioned that working here too
- 16:25:14 [bobP]
- ... development has been fairly good. We have a local copy, have stripped out immediate utility data
- 16:25:33 [bobP]
- ... can share by email. Plan is to load local DB here.
- 16:25:45 [bobP]
- ... how to expose for the rest of the group?
- 16:26:28 [bobP]
- Matthias: Problem may be larger? Ensure the scalability of the data
- 16:26:43 [bobP]
- Michel: Agree, going to make a note of this.
- 16:27:25 [bobP]
- BobF: Institutional guidlines are strict on external servers.
- 16:27:36 [bobP]
- ... approval to stand up an instance here
- 16:28:38 [bobP]
- Michel: You can load it at Carleton
- 16:29:06 [bobP]
- ... SWObjects, expose sparql endpoint?
- 16:29:24 [bobP]
- Scott: Iker has down exploratory work w d2r
- 16:29:56 [bobP]
- ... idea was to take specific data w dbSNP and access via sql mapped in swobjects
- 16:30:08 [bobP]
- ... not trying to expose everything
- 16:30:24 [bobP]
- ... Iker found that infra was the problem
- 16:30:45 [bobP]
- (I think he found that it was too big etc)
- 16:31:05 [Zakim]
- -??P7
- 16:31:38 [bobP]
- Iker: Did mapping w d2r, swobjects. Used a subset, here is the link
- 16:31:52 [iker]
- https://cgsmd.isi.edu/dbsnpq/
- 16:32:46 [bobP]
- ... first we should generate a subset of dbSNP that we will use. Huge
- 16:33:13 [bobP]
- Michel: BobF has done this part.
- 16:33:44 [bobP]
- BobF: Pulled out only those fields that are relevant. Excluded historical groupings of the RFs themselves
- 16:34:10 [bobP]
- ... contig mappings omitted. Just took SNPs, genes, classification etc
- 16:34:19 [iker]
- q+
- 16:34:21 [bobP]
- ... four different tables, but can extend from there.
- 16:34:49 [iker]
- SNP table at https://cgsmd.isi.edu/dbsnpq/downloads.php
- 16:34:57 [bobP]
- Michel: No contigs?
- 16:35:06 [iker]
- this table collects more or less the information Bob is talking about
- 16:35:13 [bobP]
- BobF: Yes, no contigs. Each build has its own mappings
- 16:35:31 [bobP]
- ... pulling in contig maps would imply pulling in the build data too
- 16:35:50 [bobP]
- ... but we can associate RS numbers w chromosomes
- 16:36:12 [bobP]
- ... also there will be some ambiguity in some
- 16:36:35 [bobP]
- ... not QC, just pulled data
- 16:36:56 [bobP]
- ... snp, chrome, validation info, locus id, proximity info, etc
- 16:37:04 [bobP]
- ... gene symbol with local ID
- 16:38:13 [bobP]
- Michel: (asking...)
- 16:38:35 [bobP]
- ... rewrite DB relations to swobjects forms
- 16:38:48 [bobP]
- Matthias: SWObjects in addition?
- 16:39:01 [iker]
- q+
- 16:39:45 [bobP]
- Michel: (implement to graph pattern - didn't quite get)
- 16:41:17 [bobP]
- Iker and Michel talking DB exchange
- 16:43:53 [bobP]
- Iker: ...Is ethnicity a requirement?
- 16:44:41 [bobP]
- Michel: Depends...longer term thing is to predict, related to pop genetics
- 16:45:16 [bobP]
- ScottScott: Scenario, have SNP for a particular patient
- 16:45:38 [bobP]
- ... might hapmap to see freq on pop. Pre-warfarin biomarker research
- 16:46:04 [bobP]
- ... warfarin, you already know whether patient has SNP or not.
- 16:46:57 [iker]
- could you provide a link to the wiki page of this work?
- 16:47:08 [bobP]
- Michel: Get trimmed down version of dbSNP up and running.
- 16:47:45 [bobP]
- Michel: Matthias' scalability issue. Would like to collaborate on a framework
- 16:48:50 [mscottm]
- @iker: we discussed that for initial versions of the paper so there is a wiki page and a few relevant google docs such as an article draft.
- 16:48:57 [bobP]
- Scott: Sustainability for HCLS, should have agree to have common method of methods, scripts, versions
- 16:49:22 [bobP]
- (that was Michel!)
- 16:49:36 [iker]
- @Scott, I am talking about the wiki page where we added the SPARQL queries of dbSNP I cannot find it
- 16:49:40 [bobP]
- Michel: Resources need to be kept up to date
- 16:50:25 [mscottm]
- for iker (in case you haven't seen this): http://www.w3.org/wiki/HCLSIG/TranslationalMedicine/pharmacogenomics
- 16:50:39 [bobP]
- Matthias: Miriam, identifiers.org, etc
- 16:50:52 [bobP]
- (that was Michel again!)
- 16:51:16 [iker]
- @Scott, maybe I am wrong but I think there was another wiki page with the SPARQL queries, I will check, thanks anyway
- 16:52:17 [bobP]
- Michel: Will try to take the medicine, on representation of pgx data.
- 16:52:55 [mscottm]
- @iker: maybe one of the google docs? Not this one (the modeling): https://docs.google.com/drawings/d/1OaqlmiTbIAkgHM6kCV-4xbRddHJiN1YZkCWdNHvwiGI/edit?authkey=CIT4rcwM&hl=en_US&authkey=CIT4rcwM
- 16:53:37 [matthias_samwald]
- maybe you meant http://www.w3.org/wiki/HCLSIG/pharmacogenomics_2011 ?
- 16:53:41 [Zakim]
- -??P26
- 16:54:22 [adrien]
- I wish you all a very good day.
- 16:54:34 [Zakim]
- - +33.3.83.59.aabb
- 16:54:51 [bobP]
- Scott: One of Bob's Mayo people has project, sounds uncannily similar
- 16:55:07 [mscottm]
- http://informatics.mayo.edu/LCD/
- 16:56:12 [bobP]
- Scott: Project talks about LD, linking clinical records etc
- 16:56:40 [bobP]
- ... sent questions to LOD group, so contacted him
- 16:57:22 [bobP]
- ... can Mayo take advantage, and we can take advantage to get use cases and data
- 16:58:40 [bobP]
- BobF: Some of the work is what is offerring here
- 17:02:05 [Zakim]
- -Tony
- 17:02:09 [Zakim]
- -Bob_Powers
- 17:02:10 [Zakim]
- - +62427aaaa
- 17:02:14 [Zakim]
- -??P55
- 17:02:16 [Zakim]
- -??P22
- 17:02:18 [Zakim]
- SW_HCLS(TMO)11:00AM has ended
- 17:02:20 [Zakim]
- Attendees were +62427aaaa, Bob_Powers, +33.3.83.59.aabb, Tony
- 17:02:42 [bobP]
- rrsagent, draft minutes
- 17:02:42 [RRSAgent]
- I have made the request to generate http://www.w3.org/2011/10/06-hcls2-minutes.html bobP
- 17:02:58 [bobP]
- rrsagent, make logs public
- 18:45:41 [Zakim]
- Zakim has left #hcls2
- 20:55:48 [matthias_samwald]
- matthias_samwald has joined #hcls2