RIDER: 3D RNA Inverse Design with Reinforcement Learning-Guided Diffusion

Tianmeng Hu, Yongzheng Cui, Biao Luo, Ke Li

Keywords: RNA Inverse Design, Reinforcement Learning, RNA Structure

TL;DR: We propose an RL-guided diffusion model for RNA inverse design that directly optimizes 3D structural similarity, generalizes across predictors, and achieves over 100% gains over baselines while generating novel sequences.

Abstract: The inverse design of RNA three-dimensional (3D) structures is crucial for engineering functional RNAs in synthetic biology and therapeutics. While recent deep learning approaches have advanced this field, they are typically optimized and evaluated using native sequence recovery, which is a limited surrogate for structural fidelity, since different sequences can fold into similar 3D structures and high recovery does not necessarily indicate correct folding. To address this limitation, we propose RIDER, an RNA Inverse DEsign framework with Reinforcement learning that directly optimizes for 3D structural similarity. First, we develop and pre-train a GNN-based generative diffusion model conditioned on the target 3D structure, achieving a $9\\%$ improvement in native sequence recovery over state-of-the-art methods. Then, we fine-tune the model with an improved policy gradient algorithm using four task-specific reward functions based on 3D self-consistency metrics. Experimental results show that RIDER improves structural similarity by over $100\\%$ across all metrics and discovers designs that are distinct from native sequences.

Primary Area: applications to physical sciences (physics, chemistry, biology, etc.)

Submission Number: 12386