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

TL;DR

RIDER is a reinforcement learning-guided diffusion framework that directly optimizes RNA 3D structural similarity, significantly improving structural fidelity over previous methods.

Contribution

It introduces a novel diffusion-based generative model conditioned on 3D structures and employs reinforcement learning for direct structural optimization.

Findings

Achieved a 9% improvement in native sequence recovery over state-of-the-art methods.

Improved structural similarity by over 100% across all metrics.

Discovered RNA designs that are structurally similar but sequence-distinct from native 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.