Deep Learning Framework for RNA Inverse Folding with Geometric Structure Potentials

TL;DR

This paper presents a deep learning framework combining GVP layers and Transformers to improve RNA inverse folding, achieving state-of-the-art results in sequence and structural accuracy across diverse RNA datasets.

Contribution

Introduces a novel deep learning model integrating geometric and sequence information for RNA design, outperforming existing methods on multiple benchmarks.

Findings

Achieves higher sequence recovery and TM-score than previous methods.

Demonstrates strong generalization across RNA families using Rfam validation.

Refolded sequences closely match native structures with AlphaFold3.

Abstract

RNA's diverse biological functions stem from its structural versatility, yet accurately predicting and designing RNA sequences given a 3D conformation (inverse folding) remains a challenge. Here, I introduce a deep learning framework that integrates Geometric Vector Perceptron (GVP) layers with a Transformer architecture to enable end-to-end RNA design. I construct a dataset consisting of experimentally solved RNA 3D structures, filtered and deduplicated from the BGSU RNA list, and evaluate performance using both sequence recovery rate and TM-score to assess sequence and structural fidelity, respectively. On standard benchmarks and RNA-Puzzles, my model achieves state-of-the-art performance, with recovery and TM-scores of 0.481 and 0.332, surpassing existing methods across diverse RNA families and length scales. Masked family-level validation using Rfam annotations confirms strong generalization beyond seen families. Furthermore, inverse-folded sequences, when refolded using AlphaFold3, closely resemble native structures, highlighting the critical role of geometric features captured by GVP layers in enhancing Transformer-based RNA design.