Join FDA/PhUSE Working Groups to Help Harness the Power of Computational Science

Abstract

Consumers are often pleased by the accuracy with which online shopping websites are able to recommend new products they may be interested in. For example, how did the favorite shopping website of your wine-enthusiast neighbor, who recently purchased a highly durable dog crate for her separation anxiety–stricken canine, know that she would also be interested in 2 items not typically bought together: a do-it-yourself wine cork bulletin board kit and canine calming herbal supplements? Amazon.com, Pandora Radio, and Netflix, to name a few, have developed innovative means to weed through and translate the heaps of data generated from a consumer’s search and ‘‘purchase’’ history into precise product recommendations. Advances in information technology that have been extremely successful for consumer information delivery have also provided effective solutions to data-heavy projects in translational science, genomics, and business analytics. These advances have engendered fields and businesses that regulatory stakeholders can leverage against the mountains of data that accompany each medical product application in order to speed review time, reduce and refine clinical trial requirements, pinpoint postmarket safety signals, and facilitate regulatory science efforts, to name a few potential benefits. Increasingly complex informatics and data challenges must first be addressed in order to harness the potential of medical product development data. Aggregation and curation of data collected from various sources, in various forms, are heavy tasks in existing scientific environments. Data standards and associated tools can make the job of transforming data into knowledge a fast, efficient, and highly repeatable process. The first step, increasing the accessibility of data with standards, is only an enabling step. The full benefit of standardized data can only be actualized when combined with systems, tools, and procedures that allow the user to easily find answers to the scientific and regulatory questions that they face. The FDA and Pharmaceutical Users Software Exchange (PhUSE) have teamed up to tackle these challenges using collaboration, crowd sourcing, and innovation, a strategy similar to one used in 2011 by scientists who were unable to solve the crystal structure of a key retroviral protein. These innovative scientists cast a wide net in order to reach a solution by challenging nonscientist gamers to model the protein’s structure using their intuition and a protein folding game. In 3 weeks, gamers were able to accomplish what biochemists had been unable to achieve for more than a decade and provided the field with new insights for the design of antiretroviral drugs. Similar examples are now becoming more prevalent as we try to harness technical and human resources to address complex scientific and health-related issues. Earlier this year, the FDA and PhUSE cast their own wide net to solve common informatics needs and challenges by calling on all interested parties from across the pharmaceutical industry, government, and technology sectors to bring their innovative ideas to the Computational Science Symposium held on March 19 and 20, 2012. The symposium, cohosted by PhUSE and the FDA’s Center for Biologics Evaluation and Research (CBER) and Center for Drug Evaluation and Research (CDER), was designed to create a space where experts could work collaboratively to identify issues, discuss possible solutions, and ultimately devise and execute solutions by participating in any of the 6 working groups. Working groups were created based on the priorities of CBER and CDER and on interests of other stakeholders identified at previous meetings. All working groups have multiple representatives from both the FDA and industry and will hold virtual meetings throughout the year to address the objectives and next steps generated during the symposium. All working groups are