Journal Article

Determining structures of RNA conformers using AFM and deep neural networks

Rein, Alan, Schwieters, Charles D., Degenhardt, Maximilia F. S., Lee, Yun-Tzai, Degenhardt, Hermann F., Bhandari, Yuba R., Wingfield, Paul T., Wang, Yun-Xing, Ding, Jienyu, Heinz, William F., Zhang, Jinwei, Watts, Norman R., Yu, Ping, Stagno, Jason R.

2025

Overview

Much of the human genome is transcribed into RNAs 1 , many of which contain structural elements that are important for their function. Such RNA molecules—including those that are structured and well-folded 2 —are conformationally heterogeneous and flexible, which is a prerequisite for function 3 , 4 , but this limits the applicability of methods such as NMR, crystallography and cryo-electron microscopy for structure elucidation. Moreover, owing to the lack of a large RNA structure database, and no clear correlation between sequence and structure, approaches such as AlphaFold 5 for protein structure prediction do not apply to RNA. Therefore, determining the structures of heterogeneous RNAs remains an unmet challenge. Here we report holistic RNA structure determination method using atomic force microscopy, unsupervised machine learning and deep neural networks (HORNET), a novel method for determining three-dimensional topological structures of RNA using atomic force microscopy images of individual molecules in solution. Owing to the high signal-to-noise ratio of atomic force microscopy, this method is ideal for capturing structures of large RNA molecules in distinct conformations. In addition to six benchmark cases, we demonstrate the utility of HORNET by determining multiple heterogeneous structures of RNase P RNA and the HIV-1 Rev response element (RRE) RNA. Thus, our method addresses one of the major challenges in determining heterogeneous structures of large and flexible RNA molecules, and contributes to the fundamental understanding of RNA structural biology. HORNET, a method that uses unsupervised machine learning and deep neural networks to analyse atomic force microscopy data enables structural determination of RNA molecules in multiple conformations.

• Publisher: Nature Publishing Group UK,Nature Publishing Group
• Subject: 631/337/1645/501 / 631/45/500 / 631/535/1262 / 631/57/2265 / Amino acid sequence / Artificial neural networks / Atomic force microscopy / Atomic properties / Atomic structure / Crystallography / Deep Learning / Electron microscopy / Humanities and Social Sciences / Humans / Machine learning / Methods / Microscopy / Microscopy, Atomic Force - methods / Models, Molecular / multidisciplinary / Neural networks / Neural Networks, Computer / NMR / Nuclear magnetic resonance / Nucleic Acid Conformation / Nucleotide sequence / Physiology / Protein structure / Ribonuclease P / Ribonucleic acid / RNA / RNA - chemistry / RNA - ultrastructure / Science / Science (multidisciplinary) / Signal to noise ratio / Structural members / Topography / Transfer RNA / Unsupervised learning / Unsupervised Machine Learning