Autoencoder-Based Detection of Dynamic Allostery Triggered by Ligand Binding Based on Molecular Dynamics

Abstract

Dynamic allostery on proteins, triggered by regulator binding or chemical modifications, transmits information from the binding site to distant regions, dramatically altering protein function. It is accompanied by subtle changes in side-chain conformations of the protein, indicating that the changes in dynamics, and not rigid or large conformational changes, are essential to understand regulation of protein function. Although a lot of experimental and theoretical studies have been dedicated to investigate this issue, the regulation mechanism of protein function is still being debated. Here, we propose an autoencoder-based method that can detect dynamic allostery. The method is based on the comparison of time fluctuations of protein structures, in the form of distance matrices, obtained from molecular dynamics simulations in ligand-bound and -unbound forms. Our method detected that the changes in dynamics by ligand binding in the PDZ2 domain led to the reorganization of correlative fluctuation motions among residue pairs, which revealed a different view of the correlated motions from the PCA and DCCM. In addition, other correlative motions were also found as a result of the dynamic perturbation from the ligand binding, which may lead to dynamic allostery. This autoencoder-based method would be usefully applied to the signal transduction and mutagenesis systems involved in protein functions and severe diseases.

Neural networks for protein structure and function prediction and dynamic analysis

“…Rearrangement of the correlative fluctuations by these residues led to propagation of the signals by ligand binding to the distant regions. These findings suggest that the leading and accompanying residues, as detected by the clustering of the DIO vectors, were involved in dynamic allostery (Tsuchiya et al 2019).…”

“…Therefore, a precise analysis of the changes of the dynamics, rather than the static conformational changes, is of fundamental importance for elucidating the regulation of protein function. We adopted an autoencoder, unsupervised neural network, to detect dynamic changes in the PDZ2 protein domain in human PTPN13, as triggered by binding of the peptide of RAPGEF6 (Tsuchiya et al 2019).…”

“…Our autoencoder-based method detected all 20 resides as (8) leading and (12) accompanying residues. Some of these accompanying residues were distant from the ligand, which suggests that these accompanying residues were involved not only in assisting the correlative fluctuations led by the leading residues; they were also involved in propagating the fluctuations to regions that are distant from the ligand (Tsuchiya et al 2019).…”

“…In the third section, we discuss our work on autoencoders (Tsuchiya et al 2019). An autoencoder is an unsupervised neural network that is widely used for analyzing specific features of proteins (Lemke and Peter 2019).…”

“…In the third section, we introduce our autoencoder-based method, where the autoencoder was used to detect changes in protein dynamics by ligand binding and the allosteric behavior. Along with that detection, we considered the dynamics observed only in ligandbound (holo) form during the simulations as "anomalies" from the ligand-unbound (apo) form (Tsuchiya et al 2019).…”