Scalable Emulation of Protein Equilibrium Ensembles with Generative Deep Learning
Valence Labs via YouTube
Overview
Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Watch a technical talk exploring BioEmu, a groundbreaking generative deep learning system designed for protein structure ensemble analysis. Discover how this innovative system generates thousands of statistically independent protein structure samples per hour using a single GPU. Learn about the system's novel training methods that leverage extensive protein structure data, molecular dynamics simulations, and experimental protein stability measurements. Explore how BioEmu successfully models various functional conformational changes, from cryptic pocket formation to large-scale domain rearrangements, with remarkable accuracy in relative free energy predictions (around 1 kcal/mol). Understand how this technology provides mechanistic insights into protein fold destabilization and generates experimentally-testable hypotheses, representing a significant advancement in predicting protein dynamical mechanisms that underlie biological functions.
Syllabus
Scalable Emulation of Protein equilibrium Ensembles with Generative Deep Learning
Taught by
Valence Labs