A state-of-the-art web platform for collaborative, longitudinal genome diagnostics

Summary/Abstract Undiagnosed diseases clinics and newborn intensive care unit rapid genome sequencing programs have emerged as the primary settings for diagnosing patients with complex phenotypes. Subspecialty clinics, pharmaceutical companies, and direct-to-consumer genetic testing companies are increasingly adopting these technologies for diagnosing patients, or studying disease with the goal of developing treatments. In all cases, a diagnostic, or research analysis approach based on comprehensive genomic sequencing data, and, crucially, collaborative analysis by expert teams has been adopted. The team-based approach leverages the expertize of a wide array of specialists, for example, the physician's knowledge of the patient's phenotypes and family history; the medical geneticist's knowledge of genetic diseases; the bioinformatician's and variant scientist's expertise in interpreting the potential diagnostic role of individual genetic variants; the genetic counselors expertize in synthesizing all available data, and coordinating with patients, families and treating physicians. A patient's phenotypes, and more frequently, the myriad sources of evidence used to support the diagnostic role of variants, genes, and their associations with disease evolve over time, and consequently research and diagnostic analysis is a long-term process. Existing tools focus on analysis at a single point in time, and require over-burdened genetic counselors to manually evaluate cases for changes, or miss the opportunity for diagnosis due to a lack of evidence at the time of evaluation. Here, we propose to build a commercial software system to address two pressing needs faced by consumers of genomic sequencing data. First, a complete end-to-end bioinformatic pipeline to process raw sequencing data to lists of prioritized genetic variants, and structural variant calls, which explicitly includes an automated (or manually triggered) data re-analysis pipeline to support long-term genomic care. Second, a comprehensive, versatile, and yet easy-to-use visualization platform will seamlessly display all genomic data and metadata to the analysis teams in web applications custom designed to address the needs of the diverse experts collaborating on these projects. We will evaluate and refine this product in our undiagnosed diseases clinic and our newborn intensive care unit, ideal settings to garner critical feedback and suggestions from teams representative of the initial target market for this tool.