RNADyn: A Benchmark for Generating and Understanding RNA Dynamics
The paper pairs a 2,585-trajectory RNA dynamics benchmark with a model meant to generate trajectories and extract dynamics fingerprints from one conformer.
RNADynBench uses quality-controlled 100-ns all-atom molecular dynamics trajectories with leakage-controlled splits. RNADynNet shares a backbone across trajectory generation and dynamics representation learning, adding coordinate denoising, single-frame-to-trajectory alignment, and physical grounding. The authors report RMSF correlations of 0.875 and 0.766 across two test sets, including a high-flexibility challenge set. ArXiv · AI/CL/LG's note
RNADynBench uses quality-controlled 100-ns all-atom molecular dynamics trajectories with leakage-controlled splits. RNADynNet shares a backbone across trajectory generation and dynamics representation learning, adding coordinate denoising, single-frame-to-trajectory alignment, and physical grounding. The authors report RMSF correlations of 0.875 and 0.766 across two test sets, including a high-flexibility challenge set. ArXiv · AI/CL/LG's note
score 5