Partners

Abstract:
RNA is a flexible biopolymer that adopts diverse conformations while forming structural motifs essential for its function. Classical RNA force fields often show limited transferability and inefficient sampling of transitions between stable states, particularly in moderately large RNA. To address these limitations, quantum-informed machine learning (ML) potentials have recently emerged as a promising alternative, offering improved accuracy and transferability relative to classical force fields. Here, we assess ML potentials for exploring RNA conformations using the adenine–adenine dinucleoside monophosphate (ApA) dimer, a fundamental RNA building block. We generated an extensive quantum-mechanical (QM) dataset of physicochemical properties for ApA conformations obtained from temperature replica exchange molecular dynamics (TREMD) simulations. Despite its small size, the ApA dimer exhibits a complex energetic landscape with six well-defined conformational clusters in which quantum effects and solvent-mediated interactions play a crucial role. Using this dataset, we parametrized ML potentials based on the equivariant MACE architecture and informed by both ab initio and semiempirical property data. The resulting potentials reproduce key conformational features of the ApA system, including base stacking, sugar geometry, and backbone flexibility, and provide broader coverage of structural transitions than the general-purpose SO3LR and MACE-POLAR-1 models. These findings underscore the importance of comprehensive QM datasets for RNA building blocks to support the structural and energetic characterization of RNA complexes and emphasize the need for robust and efficient validation metrics for ML potentials.

Keywords:
Machine Learning, RNA, Free Energy, Conformational Changes , MD, Genetics, Cluster Analysis, Molecular Mechanics

Abstract:
Biomolecular simulations played a crucial role in advancing our understanding of the complex dynamics in biological systems with applications ranging from drug discovery to the molecular characterization of virus-host interactions. Despite their success, biomolecular simulations face inherent challenges due to the multiscale nature of biological processes, which involve intricate interactions across a wide range of length scales and timescales. All-atom (AA) molecular dynamics provides detailed insights at atomistic resolution, yet it remains limited by computational constraints, capturing only short timescales and small conformational changes. In contrast, coarse-grained (CG) models extend simulations to biologically relevant time and length scales by reducing molecular complexity. However, CG models sacrifice atomic-level accuracy, making the parameterization of reliable and transferable potentials a persistent challenge. This review discusses recent advancements in machine learning (ML)-driven biomolecular simulations, including the development of ML potentials with quantum-mechanical accuracy, ML-assisted backmapping strategies from CG to AA resolutions, and widely used CG potentials. By integrating ML and CG approaches, researchers can enhance simulation accuracy while extending time and length scales, overcoming key limitations in the study of biomolecular systems.

Keywords:
Machine Learning, Coarse graining, Molecular Simulations, Proteins, MACE, Neural Network, Back-mapping, all-atom MD