Neural relational inference to learn long-range allosteric interactions in proteins from molecular dynamics simulations

• Published in: Nature communications 2022-03, Vol.13 (1), p.1661-16, Article 1661
• Main Authors: Zhu, Jingxuan, Wang, Juexin, Han, Weiwei, Xu, Dong
• Format: Article
• Language: English
• Subjects: 119/118; 631/114/2397; 631/45/173; Allosteric properties; Allosteric Regulation; Allosteric Site; Amino acids; Biological activity; Catalytic Domain; Coders; Computer applications; Deep learning; Encoders-Decoders; Free energy; Graph neural networks; Humanities and Social Sciences; Inference; Molecular dynamics; Molecular Dynamics Simulation; multidisciplinary; Mutation; Neural networks; Pin1 protein; Proteins; Proteins - chemistry; Residues; Science; Science (multidisciplinary); Simulation; Superoxide dismutase
• ISSN: 2041-1723; 2041-1723
• DOI: 10.1038/s41467-022-29331-3

Protein allostery is a biological process facilitated by spatially long-range intra-protein communication, whereby ligand binding or amino acid change at a distant site affects the active site remotely. Molecular dynamics (MD) simulation provides a powerful computational approach to probe the allosteric effect. However, current MD simulations cannot reach the time scales of whole allosteric processes. The advent of deep learning made it possible to evaluate both spatially short and long-range communications for understanding allostery. For this purpose, we applied a neural relational inference model based on a graph neural network, which adopts an encoder-decoder architecture to simultaneously infer latent interactions for probing protein allosteric processes as dynamic networks of interacting residues. From the MD trajectories, this model successfully learned the long-range interactions and pathways that can mediate the allosteric communications between distant sites in the Pin1, SOD1, and MEK1 systems. Furthermore, the model can discover allostery-related interactions earlier in the MD simulation trajectories and predict relative free energy changes upon mutations more accurately than other methods. Here, the authors apply a neural relational inference model to infer dynamic networks of interacting residues in protein molecular dynamics simulations. The model can predict allosteric communication pathways and relative free energy changes upon mutations.