Abstract: In this talk we provide an overview of how artificial intelligence / machine learning techniques are being used in life sciences research, biomedicine, and drug discovery. We highlight important specific applications of AI/ML techniques, across domains such as medical imaging, genomics from experimental data interpretation to understanding the genome, small molecule drug discovery. We also discuss recent advances using deep learning techniques to model protein sequences and structures, from basic scientific research to the design of novel proteins for chemical and therapeutic applications. We end with a brief overview of some open challenges in the field.
Bio: Mark DePristo is the Founder and CEO of BigHat Biosciences, an early-stage Bay Area startup applying AI/ML techniques to the design and optimization of next-generation antibodies. From 2016-2019, Mark DePristo founded and then led the Genomics team in Google Brain, which applies deep learning in TensorFlow to genomics problems to create tools such as DeepVariant and Nucleus and research like A deep learning approach to pattern recognition for short DNA sequences and Using deep learning to annotate the protein universe. Before joining Google he was Vice President of Informatics at SynapDx, a Google Ventures-backed startup developing a blood-based test for Autism. As Co-Director of Medical and Population Genetics at the Broad Institute from 2008-2013 Mark created and led the team that developed the GATK, the gold standard software for processing next-generation DNA sequencing data. He has a BA in Computer Science and Math from Northwestern University, a PhD in Biochemistry from the University of Cambridge as a Marshall Scholar, and did a postdoc at Harvard University in evolutionary biology. Dr. DePristo's academic articles are widely published and have received more than 58,000 citations.