Determining structures of RNA conformers using AFM and deep neural networks.

Area of Science:

• Molecular Biology
• Structural Biology
• Biophysics

Background:

• Many functional RNAs are conformationally heterogeneous and flexible, posing challenges for traditional structure determination techniques like NMR and cryo-EM.
• Existing computational methods, such as AlphaFold, are not directly applicable to RNA due to the lack of comprehensive structure databases and clear sequence-structure correlations.
• Accurate determination of three-dimensional RNA structures is crucial for understanding their biological functions.

Purpose of the Study:

• To develop a novel method for determining the three-dimensional topological structures of heterogeneous RNA molecules.
• To overcome the limitations of existing methods in elucidating the structures of large, flexible, and conformationally diverse RNAs.
• To provide a new tool for advancing RNA structural biology and understanding RNA's role in biological systems.

Main Methods:

• Development of the Holistic RNA structure determination method using Atomic Force Microscopy, unsupervised machine learning, and deep neural networks (HORNET).
• Utilizing atomic force microscopy (AFM) to capture high-resolution images of individual RNA molecules in solution.
• Applying unsupervised machine learning and deep neural networks to analyze AFM data and determine 3D topological structures.

Main Results:

• HORNET successfully determined multiple heterogeneous structures of RNase P RNA and the HIV-1 Rev response element (RRE) RNA.
• The method demonstrated high signal-to-noise ratio, enabling the capture of distinct conformations in large RNA molecules.
• Validation was performed using six benchmark cases, confirming the method's accuracy and applicability.

Conclusions:

• HORNET provides a robust solution for determining the heterogeneous structures of large and flexible RNA molecules.
• This advancement addresses a significant challenge in RNA structural biology, enabling deeper insights into RNA function.
• The method contributes to the fundamental understanding of RNA structural dynamics and their implications in biological processes.