About the Webinar
The Genomics team in Google AI develops DeepVariant, a deep learning-based variant caller distinguished for its high accuracy. Because deep learning methods learn features from training data, DeepVariant can be quickly retrained to new data types. In this talk, we explore how DeepVariant was extended for PacBio CCS data, PCR positive whole genome sequencing, exomes, BGISEQ data, and non-human genomes without any additional code. Further, because DeepVariant learns features without direct instruction, we will investigate examples where DeepVariant was able to discover features in the data missed by humans.
About the Presenter
Andrew Carroll holds a Ph.D. in Molecular Biology from Stanford University, and a Bachelor’s degree in Physics from the University of Virginia. He currently works as part of the Genomics team in Google AI, who work to develop methods to help understand genomic data and combine it with non-genomic biomedical data. Andrew shapes product strategy to drive adoption of the technology and ensure its connection to high-value application. His role has a strong focus on genomics community engagement, collaboration, and partnership.
Prior to Google, Andrew was Chief Scientific Officer at DNAnexus. There he grew and led a team of bioinformaticians who supported many of the first large-scale genomics projects, such as the CHARGE Consortium, Regeneron-Gesinger and Regeneron-UKBiobank cohorts, the 3000 Rice Genomes Project, PrecisionFDA, and the St. Jude Pediatric Cancer Cloud.