Memory kernel minimization based neural networks for discovering slow dynamic modes of molecular kinetics | Poster Board #1420

Identifying collective variables (CVs) that accurately capture the slowest timescales of protein conformational changes is crucial to comprehend numerous biological processes. In this work, we develop a novel algorithm, the Memory kErnel Minimization based Neural Networks (MEMnets), that accurately identifies the slow CVs of biomolecular dynamics. MEMnets algorithm is distinct from popular deep-learning approaches (such as VAMPnets and SRVNet) that assume Markovian dynamics. Instead, MEMnets is built on the generalized master equation theory, which incorporates non-Markovian dynamics by encoding them in a memory kernel. MEMnet’s key innovation is the development of a novel loss function that incorporates the integrals of memory kernels. By incorporating a term for the integrals of memory kernels into the loss function, we demonstrate that our MEMnets algorithm can effectively identify the slow CVs involved in the folding of FIP35 WW-domain with high accuracy and elucidate the true time scales of the slowest dynamics. Furthermore, we test MEMnet’s on the clamp opening of a bacterial RNA polymerase, a much more complex conformational change (a system containing over 540K atoms), where sampling from all-atom MD simulations is limited. Our results demonstrate that MEMnets greatly outperforms SRVNet, which is based on Markovian dynamics and may result in disconnected dynamics along the identified CVs.