HCLSIG/Pharmacogenomics/Meetings/2013-05-01 Conference Call

From W3C Wiki
Meeting:      Clinical Pharmacogenomics
Date:         May 01, 2013
Time:         10:15 Eastern Time (16:15 Central European Time)
Frequency:    1st and 3rd Wednesday of the month
Conveners:    Michel Dumontier, Matthias Samwald
Dial-In #:    +1.617.761.6200 (Cambridge, MA)
VoIP address: sip:zakim@voip.w3.org
Participant   Access Code: 4257 ("HCLS")
IRC Channel:  irc.w3.org port 6665 channel #HCLS

Suggested Agenda

  • Planning and priorities meeting. Prepare for this meeting by jotting down specific outcomes you want to see in the next 3, 6, 12 months. We will discuss these and determine what specifically needs to be done to achieve them. Then we will prioritize tasks and identify who wants to work on what.
  • Submission about Genomic CDS (and other ontologies you have in the field of pharmacogenomics?) to the OWL Reaser Evaluation Workshop 2013 http://ore2013.cs.manchester.ac.uk/ (Deadline 3rd of May!!)

Meeting Minutes


Notepad for documenting 3-6-12 month plans

Matthias Samwald

3 months:

  • Paper and presentation at OWL Reasoner workshop
  • Finalize re-implementation of decision support service (behind safety-code.org) in TrOWL. Need to establish collaboration with Jeff Pan and his group!!
  • Submit paper about Genomic CDS to the journal Bioinformatics
  • Tidy up / reorganize the W3C wiki and http://www.genomic-cds.org/

6 months:

  • Have post-doc in Vienna working on this full-time!
  • Finalize overhaul and update of Genomic CDS ontology
  • Need to establish more official collaboration with PharmGKB (Michel moving to Stanford might facilitate that further)
  • More official collaboration with the group at Mayo Clinic (Bob Freimuth?)
  • Finalize grant application for additional 3-year medical PhD position -- for someone working on the biomedical details / curation / working and evaluating with doctors

12 months:

  • Genomic CDS ontology covers everything it needs to cover, connection to Rich's annotation work is continuously kept up-to-date
  • Begin partnering with external organizations (clinics, doctors, pharmacies, pharmaceutical companies) to start evaluating and dissemminating the ontology and decision support solutions (as pure research project only, since it is still undecided how best to get the software certified as a medical device). Marshfield Clinic might be a good partnering organisation for pilot studies?

Richard Boyce

3 months:

  • semantically annotated product label information for [1] published as valid Open Data Annotation and with best practice provenance (including IAA).

6 months:

  • Qualitative and possibly quantitative (e.g., task-oriented usability comparison) data on the value of integrating the semantic annotations into a prototype clinical pharmacogenomic information system.
  • Acquisition of pilot funding (~25K) for development
  • a conference paper describing the above research activities

9 months:

  • Demonstration integration of triggers and recommendations present in the Genomic CDS ontology within the UPMC system using Quest lab results and focused meds (e.g., warfarin, clopidogrel, and some psych drugs). Further qualitative inquiry.
  • Submission of a journal article describing the results from all above research activities.
  • Submission of a journal paper presenting an analysis of pharmgx statements in product including the range of content, frequency of updates, contrasts with other sources, and recommendations for clinicians, drug information compendia, and the FDA

References:

1. atomoxetine, atorvastatin, boceprevir, capecitabine, carbamazepine, carvedilol, celecoxib, cisplatin, citalopram, clobazam, clomipramine, clopidogrel, clozapine, codeine, dapsone, desipramine, dexlansoprazole, diazepam, doxepin, esomeprazole, fluorouracil, fluoxetine, flurbiprofen, fluvoxamine, iloperidone, imipramine, irinotecan, ivacaftor, mercaptopurine, metoprolol, nefazodone, nortriptyline, omeprazole, pantoprazole, paroxetine, peginterferon, alfa-2b-il28b, perphenazine, phenytoin, pimozide, pravastatin, propafenone, propranolol, protriptyline, quinidine, rabeprazole, rasburicase, rifampin, risperidone, telaprevir, terbinafine, tetrabenazine, thioguanine, thioridazine, ticagrelor, tramadol, venlafaxine, and voriconazole


Michel Dumontier

3 months:

  • describe work done this past term with curating class-level drug-drug interactions
  • curate and validate annotations for drug effects, drug indications and drug-drug indications on drug product labels (with Rich)
  • identify and develop priority datasets to be included in Bio2RDF that would benefit this task force

6 months:

  • be settled at Stanford University
  • link drug-drug interactions with pharmacogenomic interactions
  • identify and develop priority datasets to be included in Bio2RDF that would benefit this task force
  • build relationships with associated partners on RDF data publication, use and analysis

9 months:

  • identify putative animal models for pharmacogenomic outcomes

Bob Freimuth

Additional ideas for discussion and prioritization. Two projects that could overlap or be done serially. Possible publications listed in brackets.

1. Ontological genomic CDS (good start, may require a little more design and demonstration)

Short term (0-2 months):

  • review Cerner's CBO (strengths/gaps for CDS) [1]
  • discuss modeling approaches for genomic CDS (what's the trigger? integration with/maintenance of knowledge base; review existing implementations (TPP); feedback to CPIC?) [1]

Mid term (1-4 months):

  • apply ontological genomic CDS approach to key CPIC guidelines (demonstration, evaluation) [1]
  • demonstrate gene-gene-drug CDS, ensure approach will scale [2]

Long term (3-9 months):

  • consider application of gene-drug class and gene-drug-drug rules (leverages drug-drug interaction data) [2]

2. Generalization of MSC (lots of potential, need to find a high-profile community to adopt it)

Short term (0-2 months):

  • discuss generalization of med safety code approach to NGS of PGx genes
  • apply generalized MSC to a real data set (demonstration) [3]

Mid term (2-4 months):

  • consider opportunities to partner with eMERGE and/or PGRN (user base for generalized MSC)