Accelerated sampling of protein dynamics using BioEmu augmented molecular simulation

Abstract Here, we introduce a workflow that combines BioEmu generated conformational ensemble with physics based molecular simulations and Markov State Model to capture biomolecular conformational dynamics. BioEmu augmented molecular simulation approach captures active-to-inactive transitions in serine-threonine kinases and resolves how disease-causing mutations lead to population shifts among distinct metastable states. Compared to AlphaFold2 based reduced multiple sequence alignment (rMSA-AF2) approaches, BioEmu generated ensemble covers a broader conformational space and, when integrated with molecular dynamics simulations, enable Boltzmann weighted sampling of rare conformational events. Overall, this protocol provides a straightforward framework for integrating generative AI based protein emulator with statistical physics to recover Boltzmann weighted conformational ensembles at scale.