Combining All-Atom Molecular Dynamics Simulation and NMR to Analyze Conformational Ensemble of Intrinsically Disordered Proteins

Abstract

Intrinsically disordered proteins (IDPs) lack stable tertiary structures and instead populate dynamic conformational ensembles, presenting unique challenges for structural characterization. In this review, we discuss the synergistic integration of all-atom molecular dynamics (MD) simulations and nuclear magnetic resonance (NMR) spectroscopy to elucidate the structural and dynamic properties of IDPs. NMR spectroscopy provides ensemble-averaged, site-specific structural and dynamic information, though its inherently sparse data limits resolution. Conversely, MD simulations yield atomically detailed trajectories but are constrained by sampling limitations and potential force field inaccuracies. Integrating both methods, using NMR data as restraints or reweighting criteria for MD simulations, improves accuracy and provides a more complete understanding of IDP behavior. Recent advancements include statistical reweighting techniques and AI-assisted methods to enhance sampling efficiency and ensemble construction. Despite progress, challenges remain in force field accuracy and seamless data integration. Future work will focus on improving force fields, developing more dynamic data integration methods, and leveraging AI for more efficient and accurate ensemble generation.