<matthias_samwald> :(

pgx of depression use case

matthias: depression is common disease, significant burden.
... electronic representation of EHR and clinical decision support
... joanne and michel have some work in this area

<matthias_samwald> michel: yes, we had a paper in briefings in bioinformatics about early work

<matthias_samwald> michel: we looked at papers, made a spreadsheet and converted that to RDF triples.

<matthias_samwald> michel: even back then there was quite a bit of knowledge on pharmacogenomics of depression

<matthias_samwald> Richard: It is an interesting topic.

<matthias_samwald> ... U Pittsburgh has one of the leading centers studying depression in the elderly.

<matthias_samwald> ... my mentor has a few large grants in the area. Large number of nursing homes involved, large number of studies.

<matthias_samwald> ... we would have use-cases driven by these researchers

<matthias_samwald> ... treatment, interventions, estimating risk of transition to major depression

<matthias_samwald> ... other piece that is interesting: a colleague did work on biomarker discovery/genotyping for 800/900 patients

<matthias_samwald> ... of course that would require negotiation and IRB approval

my paper: http://dumontierlab.com/publications/2008_BIB_pharmacogenomics.pdf

<matthias_samwald> ... two basic branches of work: 1) pgx associated with effectiveness. limited number of pathways is affected by common antidepressants.

<matthias_samwald> 2) safety of antidepressants. this is more worked out than (1).

<bobP> http://bib.oxfordjournals.org/content/early/2009/02/24/bib.bbn056.full

<matthias_samwald> ... also there can be concerns specific to elderly

<matthias_samwald> ... i am working on statistical models about falling and other adverse events in such situations

<matthias_samwald> matthias: for the 800/900 patients, does data contain also clinical data / drug response?

jeremy: first time joining the call, what's semantic web about pharmacogenomics

matthias: we have an interest in the topic, and use semantic web to represent and reason about the underlying data
... several of us have developed different technologies: michel has linked datasets in bio2rdf, rich worked on processing structured product labels, matthias on genomic cds ontology to formally represent allele definitions and to link to guidelines to support clinical decision support
... everything is interlinked in some way

jeremy: depression is broad condition; could biomarkers be conflated with more specific sub-conditions?

matthias: what makes this interesting is that there are global mechanisms to pharmacogenomics, eg. action of drug metabolizing enzymes
... follow up with rich to find out more about his pilot study, joanne has students; matthias will identify allele variants

<matthias_samwald> (should i scribe?)

bobF: depression is a perfectly acceptable phenotype; but if we look to much into the mechanism of why drugs do/don't work, might be problematic. sweet spot is looking at genetic associations.

<matthias_samwald> okay

bobF: internally working on EHR data mining algorithms
... is that too far out scope?

matthias: interesting, but a little bit more distantly related

<JeremyCarroll> http://lists.w3.org/Archives/Public/www-archive/2013Apr/att-0002/W3C-JJC-LifeSci.pdf

jeremy on VCF and RDF

<boycer> I need to go now Please let me know when to discuss more about depression - later this month or early May. take care!

<matthias_samwald> Bye, Rich!

jeremy: synapse discovery provides solution for NGS-based diagnostics, performs data integration with semantics
... task to work with VCF data
... first approach, works, not optimized
... 15B triples for chromosome 7 from 1000 genomes

<ericP> [slide 12]

jeremy: the vcf file has a metadata header, followed by data for each variant

<ericP> [slide 13]

jeremy: then per alternative allele per sample, variant call

<ericP> [slide 17]

<ericP> [slide 19]

jeremy: is the data completely all useful?
... if we prepared some queries up front, we could hide the complexity
... optimize for those queries

<bobP> Does this actually give a likelihood for the haplotype (whether correct or not)?

<JeremyCarroll> http://www.1000genomes.org/wiki/Analysis/Variant%20Call%20Format/vcf-variant-call-format-version-41

<JeremyCarroll> "QUAL phred-scaled quality score for the assertion made in ALT. i.e. -10log_10 prob(call in ALT is wrong). If ALT is ”.” (no variant) then this is -10log_10 p(variant), and if ALT is not ”.” this is -10log_10 p(no variant). High QUAL scores indicate high confidence calls. Although traditionally people use integer phred scores, this field is permitted to be a floating point to enable higher resolution for low confidence calls if desired. [CUT]

<matthias_samwald> michel: use of rdf:seq and rdf:bag is not ideal

matthias: what's the plan for making the scripts available?

<matthias_samwald> http://www.genomic-cds.org/ont/genomic-cds.owl

<JeremyCarroll> regrets for next time

This is scribe.perl Revision: 1.137  of Date: 2012/09/20 20:19:01  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

No ScribeNick specified.  Guessing ScribeNick: michel
Inferring Scribes: michel
Default Present: Bob_Powers, +1.415.586.aaaa, michel, JeremyCarroll, +1.206.371.aabb, matthias_samwald, richard_boyce, iker, +1.507.269.aaee, EricP, EricP.a
Present: Bob_Powers +1.415.586.aaaa michel JeremyCarroll +1.206.371.aabb matthias_samwald richard_boyce iker +1.507.269.aaee EricP EricP.a

WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth

Got date from IRC log name: 03 Apr 2013
Guessing minutes URL: http://www.w3.org/2013/04/03-HCLS-minutes.html
People with action items: