IsoNet2 determines cellular structures at submolecular resolution without averaging

Research Square 2026 Preprint Open access

Read full text
Authors: Z. Hong Zhou, Yun-Tao Liu, Hongcheng Fan, Jonathan Jih, Liam Tran, Xiaoying Zhang
DOI: 10.21203/rs.3.rs-8363763/v1
Cited by: 2

Abstract

Abstract We introduce IsoNet2 , an end-to-end self-supervised deep-learning method that directly reconstructs high-quality 3D densities from cryogenic electron tomograms. A unified network simultaneously performs denoising, contrast transfer function correction, and missing-wedge restoration, achieving ~ 20 Å resolution without averaging. A feature-rich GUI enables rapid, dataset-specific fine-tuning for end-users. IsoNet2 resolves domain organization in HIV capsid proteins, tRNA occupancy in individual ribosomes, and in situ architectures of mitochondrial respiration-related complexes, enabling atomic-level interpretation of cellular environments.

Fields of Study & Subjects

Advanced Electron Microscopy Techniques and Applications Machine Learning in Bioinformatics ATP Synthase and ATPases Research

Publication details

Journal / source
Research Square
Publisher
Springer Science and Business Media LLC
Publication date
21 January 2026
Language
EN
Document type
Preprint