Boosting Protein Graph Representations through Static-Dynamic Fusion
Preprint
Abstract
Machine learning for protein modeling faces significant challenges due to proteins’ inherently dynamic nature, yet most graph-based machine learning methods rely solely on static structural information. Recently, the growing availability of molecular dynamics trajectories provides new opportunities for understanding the dynamic behavior of proteins; however, computational methods for utilizing this dynamic information remain limited. We propose a novel graph representation that integrates both static structural information and dynamic correlations from molecular dynamics trajectories, enabling more comprehensive modeling of proteins. By applying relational graph neural networks (RGNNs) to process this heterogeneous representation, we demonstrate significant improvements over structure-based approaches across three distinct tasks: atomic adaptability prediction, binding site detection, and binding affinity prediction. Our results validate that combining static and dynamic information provides complementary signals for understanding proteinligand interactions, offering new possibilities for drug design and structural biology applications.Keywords
This publication has 36 references indexed in Scilit:
- Engineering of Halide Methyltransferase BxHMT through Dynamic Cross‐Correlation Network AnalysisAngewandte Chemie International Edition, 2024
- Encoding protein dynamic information in graph representation for functional residue identificationCell Reports Physical Science, 2022
- Time Evolution of the Millisecond Allosteric Activation of Imidazole Glycerol Phosphate SynthaseJournal of the American Chemical Society, 2022
- Development of Range-Corrected Deep Learning Potentials for Fast, Accurate Quantum Mechanical/Molecular Mechanical Simulations of Chemical Reactions in SolutionJournal of Chemical Theory and Computation, 2021
- Modeling Relational Data with Graph Convolutional Networks2018
- Forging the Basis for Developing Protein–Ligand Interaction Scoring FunctionsAccounts of Chemical Research, 2017
- PTRAJ and CPPTRAJ: Software for Processing and Analysis of Molecular Dynamics Trajectory DataJournal of Chemical Theory and Computation, 2013
- Atomistic Kinetic Model for Population Shift and Allostery in BiomoleculesJournal of the American Chemical Society, 2011
- Quantifying Correlations Between Allosteric Sites in Thermodynamic EnsemblesJournal of Chemical Theory and Computation, 2009
- Full correlation analysis of conformational protein dynamicsProteins: Structure, Function, and Bioinformatics, 2007