We propose building a computational platform based on the high performance Berkeley Big Data Analytics Stack (BDAS) to support a new ecosystem of Clinical Decision Support (CDS) applications. This platform will make it faster, easier, and less expensive to develop molecular Clinical Decision Support Systems. These systems require real-time queries of globally distributed data, efficient machine learning on large genomic datasets, and must be secure, fault-tolerant and scalable. BDAS and associated technologies are designed to help us meet these challenges and are therefore ideal building blocks to help us create our computational platform. To encourage the adoption of standards for the querying and sharing of large genomic datasets, we will adapt the BDAS stack to support the standards of the Global Alliance for Genomics and Health (GA4GH).
Funding this work will help establish a production quality FOSS implementation of the important Global Alliance for Genomics and Health standards. Without such open-source implementations, a fragmented and proprietary platform ecosystem would slow down innovation as well as divert resources away from the practice of medicine.
