OpenCRAVAT Community Building for Integrated Variant Annotation Framework

Understanding the disease relevance of all genetic variants is a major goal of genomic medicine, one enabling life-changing advances in disease diagnosis, prognosis and treatment. Computational tools are essential to this goal as they integrate information from clinical studies, systems biology, animal models and machine learning. With NIH support, our team developed a system for automated cancer variant annotation and visualization that has been used to analyze ~3×109 variants for 5,000 scientific users over the past six years. CRAVAT integrates information from several computational tools developed in the Karchin Lab at Johns Hopkins University. However, variants influence thousands of diseases, and comprehensive annotation will require the combined talents of geneticists, biologists, physician-scientists, statisticians and computer scientists. Although many databases, software tools and web servers exist, it remains difficult to analyze large user-defined variant sets in the context of many resources from different institutions. To address this problem, this team developed OpenCRAVAT to provide easy access to a large catalog of resources and facilitate developer interactions. Developer contributions are similar to the Chrome store, with data made available as plug-ins and served to users as products in a store.