HCLSIG/PharmaOntology/Meetings/2011-05-12 Conference Call/Minutes
Minutes for the HCLS Translational Medicine conference call for May 12, 2011
... HCLS summary; to promote and develop the use of semweb
... HCLS charter, new one is developing
... Translational Medicine: link info, close the gap between treatment, clinical research, etc
... also identify gaps in knowledge
... paper: how to represent pgx; symptoms, signs, treatments, genetic and physiol backgrounds
... ultimately toward a predictive context
... Three new people will join us
... Dr. Fred Whipple
Fred: Specialist in SNP-based diagnostics, prognostics; breast cancer etc
... director of Genomics Initiative to educate clincians
... education gets them to understand SNPs, risk association, risk tools
... PhD Tufts; Harvard Med; Cal State; then business!
Joanne: Joanne and Fred go way back.
... to Wang Labs, in the 1970s
Michel : Dr. Robert Freimuth ... BIO: My doctoral work was in the field of pharmacogenomics (wet lab work). I have since migrated through the fields of bioinformatics and medical informatics. I have been very active in the National Cancer Institute's Cancer Biomedical Informatics Grid program, where I lead the Information Representation Working Group and its development of the Life Sciences Domain Analysis Model (LS DAM). I am also the Director of the Vocabulary Knowledge Center (VKC), which provides education and support for the NCI semantic infrastructure. Finally, I am co-lead of the Ontology Network Resource (PHONT) for the Pharmacogenomics Research Network (PGRN).
BobF : Junior investigator, working thru bioinf at Mayo
... NCI caBIG program; have worked on data standards; leading inf-representation WB
... Second project is PGx research network; recently funded to bring standardization
... interested in progress in this group
Michel: Coincides well with our work; timely
BobF: Have known pharmGKB group; have established collab w them
... help them to incorporate mappings; rolling towards semweb, expose in RDF
... application point for this work
... aware of HCLS; convergences with my projects; number of people have such experience
Iker: Support personalized med by using sem tech
... aim to relate patient data w available info;
Michel: is there predictive work also?
Iker: Prototype does not predict; will be working on pgx; we are close to the market
Michel: What about the demo that uses the TMKB
Joanne: TMO / TMKB Journal of Biomedical Semantics Link: http://www.jbiomedsem.com/content/2/S2/S1/abstract Publ Date May 17, 2011 (ready on line and in PDF)
... Alzheimer knowledge sharing platform
... "How many patients experienced side effects while taking Donepezil?"
Joanne: Queries look different when they're live!
Iker: Goal is to enroll as many patients as possible
... aim to develop a platform like this one for breast cancer
Michel: take a look at http://www.patientslikeme.com/
... want to define a gathering point for patients, physicians and researchers
Joanne: Patient's names anonymized?
... Here's the link to the SMART Health App Challenge: http://www.smartplatforms.org/challenge/
... represent pgx for warfarin
... need to define patient-centered use cases
... warfarin-sensitive, -insensitive
... tabular form for dosing decision
... convert the table to RDF; then query
epichler: who has access to the papers that describe the prediction algorithms and could share those with me? - thanks
Bob: Paper from Chest: http://www.ncbi.nlm.nih.gov/pubmed/18574265
matthias: the paper compares the efficacy of a simple table that makes dosage recommendations for specific combinations of alleles with a more sophisticated algorithm.
matthias: we need to (at least) be able to represent combinations of alleles.
BobF discussing how a *1 allele defintion can change
BobF haplotypes p.164 S H1, H2 vs H7, H8, H9
BobF: Allele nomenclature can become out of date, one must map thru time
BobF: This is like a superclass
... but *not* a nice clean hierarchy
... subclass can be from multiple superclasses.
Joanne: Can we get an example to chew on off-line?
Scott: This is an important modelling issue: nesting of SNPs and dependence between them
Michel: We need to focus on paper; add comment w goog doc