Accurate RNA 3D Structure Prediction via Language Model-Augmented AlphaFold 3

Shuxian Zou, Jiayou Zhang, Bingkang Zhao, Hui Li, Le Song

Keywords: RNA 3D structure prediction; AlphaFold 3; RNA language model

Abstract: Predicting RNA 3D structure from sequence remains challenging due to the structural flexibility of RNA molecules and the scarcity of experimentally resolved structures. We ask how self-supervised RNA language models (LMs), trained on millions of RNA sequences, can best enhance AlphaFold 3 (AF3) for RNA structure prediction. Using an open-source AF3 reproduction, we run controlled experiments that fix data and hyperparameters while varying fusion position and method. We find large performance variations between fusion strategies, and without Multiple Sequence Alignment (MSA), they are generally not effective. When incorporating MSA, the most effective approach is additive fusion applied at the late stage of the conditional network, refining AF3’s single representations with RNA LM embeddings. On RecentPDB-RNA (47 newly released targets), our best model achieves an average TM-score of 0.438 and a success rate of 30\% (TM-score $\ge$ 0.6), significantly outperforming all baseline models. On 11 CASP16-RNA targets, it matches the best automated system trRosettaRNA. These results show that properly fused RNA LM features substantially advance RNA 3D structure prediction. We will release the data, code, and model weights to support open science, reproducibility, and the development of automated RNA structure prediction models.

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

Submission Number: 20867