Visualizing conformational space of functional biomolecular complexes by deep manifold learning

• Published in: bioRxiv
• Main Authors: Wu, Zhaolong, Chen, Enbo, Zhang, Shuwen, Ma, Yinping, Liu, Congcong, Chang-Cheng, Yin, Mao, Youdong
• Format: Paper
• Language: English
• Published: Cold Spring Harbor Cold Spring Harbor Laboratory Press 18.09.2021
• Subjects: Deep learning; Electron microscopy; Embedding; Intermediates; Macromolecules; Medical research
• DOI: 10.1101/2021.08.09.455739

Abstract The cellular functions are executed by biological macromolecular complexes in nonequilibrium dynamic processes, which exhibit a vast diversity of conformational states. Solving conformational continuum of important biomolecular complexes at atomic level is essential to understand their functional mechanisms and to guide structure-based drug discovery. Here we introduce a deep learning framework, named AlphaCryo4D, which enables atomic-level cryogenic electron microscopy reconstructions of conformational continuum. AlphaCryo4D integrates 3D deep residual learning with manifold embedding of energy landscapes, which simultaneously improves 3D classification accuracy and reconstruction resolution via an energy-based particle-voting algorithm. By applications of this approach to analyze five experimental datasets, we demonstrate its generality in breaking resolution barrier of visualizing dynamic components of functional complexes, in choreographing continuous inter-subunit motions and in exploring their ‘hidden’ conformational space. Our approach expands the realm of structural ensemble determination to the nonequilibrium regime at atomic level, thus potentially transforming biomedical research and therapeutic development. Competing Interest Statement The authors have declared no competing interest. Footnotes * This version adds additional tests on the non-uniform distribution in the conformational continuum using more synthetic datasets, as shown in the new Supplementary Fig. 6.