Efficient high-resolution refinement in cryo-EM with stochastic gradient descent
- 23 June 2025
- journal article
- Published by International Union of Crystallography (IUCr) in Acta Crystallographica Section D Structural Biology
- Vol. 81 (7) , 327-343
Abstract
Electron cryo-microscopy (cryo-EM) is an imaging technique that is widely used in structural biology to determine the three-dimensional structure of biological molecules from noisy two-dimensional projections with unknown orientations. As the typical pipeline involves processing large amounts of data, efficient algorithms are crucial for fast and reliable results. The stochastic gradient descent (SGD) algorithm has been used to improve the speed of ab initio reconstruction, which results in an initial, low-resolution estimation of the volume representing the molecule of interest, but has yet to be applied successfully in the high-resolution regime, where expectation–maximization algorithms achieve state-of-the-art results, at a high computational cost. In this article, we investigate the conditioning of the optimization problem and show that the large condition number prevents the successful application of gradient descent-based methods at high resolution. Our results include a theoretical analysis of the condition number of the optimization problem in a simplified setting where the individual projection directions are known, an algorithm based on computing a diagonal preconditioner using Hutchinson's diagonal estimator and numerical experiments showing the improvement in the convergence speed when using the estimated preconditioner with SGD. The preconditioned SGD approach can potentially enable a simple and unified approach to ab initio reconstruction and high-resolution refinement with faster convergence speed and higher flexibility, and our results are a promising step in this direction.Keywords
All Related Versions
Funding Information
- National Institutes of Health (R01GM136780)
- Air Force Office of Scientific Research (FA9550-21-1-0317)
- Alfred P. Sloan Foundation (FG-2023-20853)
- Defense Advanced Research Projects Agency (HR00112490485)
- Simons Foundation
This publication has 46 references indexed in Scilit:
- Time resolution in cryo-EM using a PDMS-based microfluidic chip assembly and its application to the study of HflX-mediated ribosome recyclingCell, 2024
- Integrating Molecular Models Into CryoEM Heterogeneity Analysis Using Scalable High-resolution Deep Gaussian Mixture ModelsJournal of Molecular Biology, 2023
- A molecular prior distribution for Bayesian inference based on Wilson statisticsComputer Methods and Programs in Biomedicine, 2022
- Multi-CryoGAN: Reconstruction of Continuous Conformations in Cryo-EM Using Generative Adversarial Networks2020
- Stochastic gradient descent, weighted sampling, and the randomized Kaczmarz algorithmMathematical Programming, 2015
- A Bayesian View on Cryo-EM Structure DeterminationJournal of Molecular Biology, 2012
- An estimator for the diagonal of a matrixApplied Numerical Mathematics, 2007
- Optimal Determination of Particle Orientation, Absolute Hand, and Contrast Loss in Single-particle Electron CryomicroscopyJournal of Molecular Biology, 2003
- A Maximum-Likelihood Approach to Single-Particle Image RefinementJournal of Structural Biology, 1998
- On the limited memory BFGS method for large scale optimizationMathematical Programming, 1989