Scientific AI in Action: Unlocking New Knowledge with AlphaFold 3 and AlphaGenome

About the Session

Beyond traditional physical experimentation, computer simulation, and statistical analytics, artificial intelligence has emerged as a fundamental new modality of science—one that does not merely optimize human workflows, but actively makes breakthroughs, generates novel theories, and creates new scientific knowledge. Hosted by the Google Cloud Advanced Computing Community, this online session showcases AlphaFold 3 and AlphaGenome as prime exemplars of this shift. These systems are autonomous engines of discovery that decode the complex spatial, physical, and regulatory laws of nature, transforming how we understand and interact with the natural world.

Healthcare and life sciences (HCLS) researchers and practitioners should tune in to see how these predictive engines are directly generating actionable biological hypotheses. By exploring how AlphaFold 3 predicts 3D biomolecular interactions and how AlphaGenome maps 1-million-base-pair sequence-to-function pathways, HCLS professionals will learn how to transition from descriptive biology to predictive medicine. This session will demonstrate how to leverage these models to discover novel therapeutic targets, predict disease-causing mutations, and design synthetic biological structures, fundamentally compressing years of speculative laboratory research into immediate, high-fidelity discoveries.

Ultimately, scientists, engineers, and technical developers from all fields should tune in to understand how their own domains can be advanced by this new paradigm. By analyzing how Google DeepMind engineered these biological models using generative diffusion and multi-modal transformers, cross-disciplinary builders will gain a clear architectural blueprint. You will walk away with the practical knowledge needed to apply these same AI design principles to represent physical laws, run automated simulations, and build your own domain-specific, AI-powered discovery applications.