# DRfold2 is a deep learning-based tool that enables efficient and accurate RNA structure prediction

## TL;DR

DRfold2 is a new deep learning tool that improves RNA structure prediction accuracy and efficiency using advanced language models and denoising techniques.

## Contribution

DRfold2 introduces a novel pre-trained RNA Composite Language Model and denoising structure module for more accurate RNA structure prediction.

## Key findings

- DRfold2 outperforms existing methods in RNA structure prediction across multiple benchmarks.
- The RCLM captures co-evolutionary patterns, significantly improving contact prediction precision by over 100%.
- Integration with AlphaFold3 further enhances prediction accuracy.

## Abstract

RNA structures are essential for understanding their biological functions and developing RNA-targeted therapeutics. However, accurate RNA structure prediction from sequence remains a crucial challenge. We introduce DRfold2, a deep learning framework that integrates a novel pre-trained RNA Composite Language Model (RCLM) with a denoising structure module for end-to-end RNA structure prediction. Based solely on single sequence, DRfold2 achieves superior performance in both global topology and secondary structure predictions over other state-of-the-art approaches across multiple benchmark tests from diverse species. Detailed analyses reveal that the improvements primarily stem from the RCLM’s ability to capture co-evolutionary pattern and the effective denoising process, with a more than 100% increase in contact prediction precision compared to existing methods. Furthermore, DRfold2 demonstrates high complementarity with AlphaFold3, achieving statistically significant accuracy gains when integrated into our optimization framework. By uniquely combining composite language modeling, denoising-based end-to-end learning, and deep learning-guided post-optimization, DRfold2 establishes a distinct direction for advancing ab initio RNA structure prediction.

Accurate RNA structure prediction remains challenging despite recent progress in computational approaches. This study develops a deep learning framework called DRFold2 that significantly enhances the accuracy of de novo RNA structure prediction by increasing contact prediction precision compared to existing methods.