Inferring a Continuous Distribution of Atom Coordinates from Cryo-EM Images using VAEs

Cryo-electron microscopy (cryo-EM) has revolutionized experimental protein structure determination. Despite advances in high resolution reconstruction, a majority of cryo-EM experiments provide either a single state of the studied macromolecule, or a relatively small number of its conformations. This reduces the effectiveness of the technique for proteins with flexible regions, which are known to play a key role in protein function. Recent methods for capturing conformational heterogeneity in cryo-EM data model it in volume space, making recovery of continuous atomic structures challenging. Here we present a fully deep-learning-based approach using variational auto-encoders (VAEs) to recover a continuous distribution of atomic protein structures and poses directly from picked particle images and demonstrate its efficacy on realistic simulated data. We hope that methods built on this work will allow incorporation of stronger prior information about protein structure and enable better understanding of non-rigid protein structures.

[1]  L. V. van Vliet,et al.  Image formation modeling in cryo-electron microscopy. , 2013, Journal of structural biology.

[2]  H. RULLGÅRD,et al.  Simulation of transmission electron microscope images of biological specimens , 2011, Journal of microscopy.

[3]  Pushmeet Kohli,et al.  Protein structure prediction using multiple deep neural networks in the 13th Critical Assessment of Protein Structure Prediction (CASP13) , 2019, Proteins.

[4]  Igor L. Medintz,et al.  FRET as a biomolecular research tool — understanding its potential while avoiding pitfalls , 2019, Nature Methods.

[5]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[6]  Julie M. Behr,et al.  A water-mediated allosteric network governs activation of Aurora kinase A. , 2017, Nature chemical biology.

[7]  Dmitry Lyumkis,et al.  Likelihood-based classification of cryo-EM images using FREALIGN. , 2013, Journal of structural biology.

[8]  Ruslan Salakhutdinov,et al.  Importance Weighted Autoencoders , 2015, ICLR.

[9]  David J. Fleet,et al.  3D Variability Analysis: Resolving continuous flexibility and discrete heterogeneity from single particle cryo-EM , 2020, bioRxiv.

[10]  Geoffrey E. Hinton,et al.  The "wake-sleep" algorithm for unsupervised neural networks. , 1995, Science.

[11]  Sjors H.W. Scheres,et al.  RELION: Implementation of a Bayesian approach to cryo-EM structure determination , 2012, Journal of structural biology.

[12]  Thorsten Wagner,et al.  SPHIRE-crYOLO is a fast and accurate fully automated particle picker for cryo-EM , 2019, Communications Biology.

[13]  E. Lindahl,et al.  Characterisation of molecular motions in cryo-EM single-particle data by multi-body refinement in RELION , 2018, bioRxiv.

[14]  Geoffrey E. Hinton,et al.  Layer Normalization , 2016, ArXiv.

[15]  Jonas Adler,et al.  Learned Primal-Dual Reconstruction , 2017, IEEE Transactions on Medical Imaging.

[16]  David J. Fleet,et al.  cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination , 2017, Nature Methods.

[17]  Simon R. Arridge,et al.  Solving inverse problems using data-driven models , 2019, Acta Numerica.

[18]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[19]  C. Villani Optimal Transport: Old and New , 2008 .

[20]  Christopher Burgess,et al.  beta-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework , 2016, ICLR 2016.

[21]  Ellen D. Zhong,et al.  CryoDRGN: Reconstruction of heterogeneous cryo-EM structures using neural networks , 2021, Nature Methods.

[22]  Diederik P. Kingma,et al.  An Introduction to Variational Autoencoders , 2019, Found. Trends Mach. Learn..

[23]  Carola-Bibiane Schönlieb,et al.  Exploiting prior knowledge about biological macromolecules in cryo-EM structure determination , 2020, bioRxiv.

[24]  Tristan Bepler,et al.  Topaz-Denoise: general deep denoising models for cryoEM and cryoET , 2019, Nature Communications.

[25]  Sonya M. Hanson,et al.  A dynamic mechanism for allosteric activation of Aurora kinase A by activation loop phosphorylation , 2017, bioRxiv.