Molecular dynamics (MD) simulation is widely used to study protein conformations and dynamics. However, the sampling suffers from being trapped in local energy minima which are hard to jump out of, leading to an inefficient sampling process and further hindering the exploration of protein movements in affordable simulation time. The advancement of deep learning provides new opportunities for protein sampling. Variational autoencoders are a class of deep learning model to learn a low-dimensional representation (referred to as the latent space) that can capture the key features of input data. Based on this characteristic, we proposed a new adaptive sampling method, latent space assisted adaptive sampling for protein trajectories (LAST), to accelerate the expansion of protein conformational space. This method comprises cycles of (i) variational autoencoders training, (ii) seed structure selection on the latent space and (iii) conformational sampling through additional MD simulations. The proposed approach is validated on four conformations in two protein systems: two metastable states of <i>E. Coli</i> adenosine kinase (ADK) and two native states of Vivid (VVD). In all four conformations, seed structures are shown to lie on the boundary of conformation distributions. Moreover, large conformational changes were observed in a shorter simulation time when compared with conventional MD (cMD) simulations in both systems. In metastable ADK simulations, LAST explored two transition paths toward two native states while cMD being trapped in an energy basin. In VVD light state simulations, LAST only takes one third of cMD simulation time to have a similar conformational space.

Iterative expansion of protein conformational space with variational autoencoder

Fall 2022 abstracts, posters, and presentations will launch in mid-September.

ACS Fall 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.