RNA structure generator captures molecule flexibility for diverse shapes
REFLEX: RNA Ensemble Generation via Flexibility-Calibrated Stochastic Bridge
Artificial Intelligence
Summary
RNA molecules can fold into many shapes to do their jobs in cells. Capturing this variety without losing detail is hard because samples can either look too similar or unrealistic. The authors present REFLEX, a computer tool that uses experimental measures of flexibility to generate many realistic RNA shapes from just the RNA sequence. REFLEX outperforms earlier methods in balancing accuracy and diversity, helping us better understand RNA behavior.
What this means in practice
- •For biotech engineers: Generate realistic ensemble models of RNA molecules to aid design of RNA-based therapeutics and biosensors.
- •For structural bioinformatics teams: Create diverse RNA structural ensembles from sequences without multiple sequence alignments for improved structure-function analysis.
Authors
Wanli Ma, Tianmeng Hu, Ke Li
Abstract
RNA molecules often function through multiple experimentally observed conformational states. A central challenge is to capture ensemble diversity without sacrificing structural fidelity: samples may collapse toward averaged representative structures or drift into unsupported conformational regions. We introduce REFLEX (RNA Ensemble generation calibrated by FLEXibility), a sequence-conditioned generator for RNA backbone-frame ensembles based on a heteroscedastic stochastic bridge on SE(3). During training, experimental B-factors calibrate residue-wise bridge widths, exposing each observed conformer through local perturbation neighborhoods with residue-specific scales while keeping constrained regions sharper. A learned flexibility condition guides the bridge-field network to recover target conformer-specific frames from these calibrated noisy states. At inference, REFLEX uses only the input sequence, requiring no MSA-subsampling heuristics. On a multi-conformer RNA benchmark, REFLEX achieves the best precision-recall trade-off for ensemble coverage across evaluation thresholds, while recovery-stress ablations indicate gains in target-frame recovery under stronger perturbations.