3D RNA Inverse Design with Reinforcement Learning-Guided Diffusion Models

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 (NSR)—a limited surrogate for structural fidelity, as different sequences can fold into similar 3D structures, and high NSR does not necessarily indicate correct folding. To address this limitation, we propose a novel two-stage framework 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 NSR over state-of-the-art methods. Then, we fine-tune the model using an improved policy gradient algorithm with four task-specific reward functions based on 3D self-consistency metrics. Experimental results show that our approach improves structural similarity by over $100\\%$ across all metrics and discovers designs that are distinct from native sequences.