RNADyn: A Benchmark for Generating and Understanding RNA Dynamics
Abstract
Ribonucleic acid (RNA) functions through conformational changes that are not fully captured by static structures. However, large-scale standardized RNA dynamics data remain limited, and existing approaches typically treat trajectory generation and dynamics understanding as separate objectives. Here, we introduce RNADynBench, a standardized RNA molecular dynamics (MD) benchmark with 2585 quality-controlled 100-ns all-atom trajectories and leakage-controlled splits. Building on RNADynBench, we develop RNADynNet, a unified model for RNA dynamics learning that uses a shared backbone for both trajectory generation and dynamics fingerprint extraction from a single conformer. It combines coordinate denoising, single-frame-to-trajectory alignment, and physical grounding to connect all-atom trajectory generation with dynamics representation learning. Physical grounding improves both generated dynamics and the physical information recoverable from these fingerprints. Across both test sets, including the high-flexibility challenge set, the generated trajectories achieve RMSF correlations of 0.875 and 0.766, while single-conformer predictions show comparable agreement with MD-derived dynamics. RNADynBench and RNADynNet together establish a benchmark and unified modeling framework for generating and understanding RNA dynamics.
Related
- Protein Titration Benchmark of CHARMM36, Amber99sb-ildn, Amber14sb, OPLS-AA/M, and Martini 2 Force Fields for Constant pH Molecular Dynamics in GROMACS.
- Does Short Molecular Dynamics Rescoring Improve the Ranking of p38α Inhibitors over Docking? A Replicate Benchmark
- Understanding the role of chromatin accessibility and compaction on loop extrusion dynamics
- Targeting the Gut Vascular Axis in Atherosclerosis: Probiotic and Phytoantioxidant Crosstalk as a Hypothesis-Generating Framework for Residual Cardiovascular Risk
- Generating Structural Ensembles of Disordered Proteins with Diffusion Models
BibTeX
@misc{huang2026rnadyn,
title = {{RNADyn: A Benchmark for Generating and Understanding RNA Dynamics}},
author = {Yiming Huang and Lennart Bastian and Hanqun Cao and Luis Vollmers and Tolga Birdal},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.03712},
eprint = {2610.03712},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.03712},
}
RIS
TY - UNPB
TI - RNADyn: A Benchmark for Generating and Understanding RNA Dynamics
AU - Yiming Huang
AU - Lennart Bastian
AU - Hanqun Cao
AU - Luis Vollmers
AU - Tolga Birdal
PY - 2026
DA - 2026/10/02
PB - arXiv
DO - 10.48550/arxiv.2610.03712
UR - https://arxiv.org/abs/2610.03712
AB - Ribonucleic acid (RNA) functions through conformational changes that are not fully captured by static structures. However, large-scale standardized RNA dynamics data remain limited, and existing approaches typically treat trajectory generation and dynamics understanding as separate objectives. Here, we introduce RNADynBench, a standardized RNA molecular dynamics (MD) benchmark with 2585 quality-controlled 100-ns all-atom trajectories and leakage-controlled splits. Building on RNADynBench, we develop RNADynNet, a unified model for RNA dynamics learning that uses a shared backbone for both trajectory generation and dynamics fingerprint extraction from a single conformer. It combines coordinate denoising, single-frame-to-trajectory alignment, and physical grounding to connect all-atom trajectory generation with dynamics representation learning. Physical grounding improves both generated dynamics and the physical information recoverable from these fingerprints. Across both test sets, including the high-flexibility challenge set, the generated trajectories achieve RMSF correlations of 0.875 and 0.766, while single-conformer predictions show comparable agreement with MD-derived dynamics. RNADynBench and RNADynNet together establish a benchmark and unified modeling framework for generating and understanding RNA dynamics.
ER -