TBI01: Panel - Dissemination of Pharmacogenomic Knowledge: Establishing a Pathway to Support Clinical Implementation
James Hoffman, St. Jude Children's Research Hospital; Michelle Whirl-Carrillo, Stanford University; Robert Freimuth, Mayo Clinic; Josh Peterson, Vanderbilt University
Pharmacogenomics is often an initial focus for the implementation of genomic medicine. To facilitate the translation of pharmacogenomic knowledge to clinical practice, authoritative resources are needed to form a knowledge base that combines genomic and medication information, which can be used to support gene-based prescribing through the electronic health record (EHR). As these resources are established, the knowledge must be represented in ways that will enable broad dissemination. This session will illustrate how three national initiatives are working together to establish a pathway to support the dissemination and clinical implementation of pharmacogenomics knowledge. Specifically, this panel will highlight the development of clinical guidelines by the Clinical Pharmacogenetics Implementation Consortium (CPIC), including its increased focus on informatics, and the dissemination of that knowledge through PharmGKB. The panel will also summarize lessons learned by the Pharmacogenetics Research Network (PGRN) Translational Pharmacogenetics Program (TPP), which identifies barriers for the implementation of pharmacogenomics, including integration with the EHR, and compares implementation approaches across diverse sites. Finally, because pharmacogenomic expertise may be concentrated in specific organizations and the technical architecture of clinical information systems varies widely, standards must be developed to share pharmacogenomic knowledge, especially genotype interpretations and prescribing recommendations.
TBI12: Late Breaking Panel - Corporate-academic Collaboration: Lessons Learned Navigating the Pitfalls and Promises of Industry Partnerships
Leonard D'Avolio, Ariadne Labs; Atul Butte, Stanford University School of Medicine; Marc Williams, Geisinger Health System
By nearly all indications the opportunity for industry-academic partnerships has never been more promising. Low cost, high throughput sequencing is creating new markets in everything from cloud computing to diagnostics and more targeted interventions. Shifting reimbursement policies are suddenly prioritizing quality and creating unprecedented opportunity for those capable of turning clinical data into knowledge. Faced with an explosion of new biological data and pressures to better understand the use of their compounds, pharmaceutical companies are seeking new relationships with clinical and academic partners. At the same time the research budgets of traditional research funding agencies are shrinking. Finally, discovery enables the opportunities to commercial results through startup companies or technology licensing. In this session a panel of experts in straddling the fine line between academia and industry speak to the challenges and opportunities presented when navigating these two very different paradigms.
TBI14: Panel - Building a Richly Connected and Highly Analyzed Genotype/Phenotype Ecosystem in a World of Data Silos
Daniel Heinze, Zato Healthcare; Scott Kahn, Illumina; Paul McOwen, Zato Healthcare; Joel Vengco, Baystate Health; Elizabeth Worthey, Medical College of Wisconsin
The ability to index, aggregate, search, navigate, analyze and share genomic and clinical data across departmental, institutional, geographic and political boundaries while maintaining security, privacy and data rights is critical to the success of translational medicine. We discuss advances in the technology of cooperative computing, information fusion and surface form ontologies with application to the translation of genomic research to clinical practice and, conversely, the application of phenotypic data to genomic research. Specifically, we describe a seminal collaboration of genomic R&D with clinical medicine as facilitated over a secure, clinically appropriate, ontology enabled, multi-centric platform for discovery across diverse genomic and clinical data sets that are stored and administered on diverse and disparate data centers and data types. This environment motivates the investigation of a variety of genotype/phenotype issues. We discuss the migration in the clinical context from sparse phenotype information to fully extracted phenotype data from the full clinical record to ontologically structured phenotype data. In the genetic research context, we discuss the migration toward a structure of deeply analyzed and organized clusters of genotype/phenotype data.
TBI20: Panel - Strategies for Sustainable Open Source Projects for Clinical and Translational Research: Lessons from the Trenches
Elizabeth Nelson, LabKey Software; Leon Rozenblit, Prometheus Research; Michael Mendis, Harvard Partners; Ben Bauman, OpenClinica; Mark Igra, LabKey Software
Theoretically, taking an open source approach can broaden the public benefits of grant-funded software projects; increase the leverage of informatics investments; draw upon a wider pool of contributors and expertise; and improve transparency, reproducibility, and extensibility. However, as Dr. Isaac Kohane has warned, open source software is "…free like a pony. You still have to feed it and clean up after it" (TEDMed, 2013). Furthermore, simply making software open source does not ensure that it will become immediately useful to others. This panel will cover practical strategies for generalizing, sustaining, and evolving open source software developed for clinical and translational research. Panel members will address sustainable business models, feasibility of grant support, implications of different open source licenses, modes of dissemination (including community norms for attracting open-source evangelists), community-building approaches, practical trade-offs, and unexpected challenges. Panelists represent open source platforms for clinical and translational research that have proven useful across multiple organizations and shown sustainability over time. Platforms include LabKey Server (http://labkey.org), RexDB (http://rexdb.org/), i2b2 (https://i2b2.org/), and OpenClinica (https://openclinica.com/).