Deep Learning for Validating and Estimating Resolution of Cryo-Electron Microscopy Density Maps †

Structures of the promoter-bound respiratory syncytial virus polymerase

The respiratory syncytial virus (RSV) polymerase is a multifunctional RNA-dependent RNA polymerase composed of the large (L) protein and the phosphoprotein (P). It transcribes the RNA genome into ten viral mRNAs and replicates full-length viral genomic and antigenomic RNAs1. The RSV polymerase initiates RNA synthesis by binding to the conserved 3'-terminal RNA promoters of the genome or antigenome2. However, the lack of a structure of the RSV polymerase bound to the RNA promoter has impeded the mechanistic understanding of RSV RNA synthesis. Here we report cryogenic electron microscopy structures of the RSV polymerase bound to its genomic and antigenomic viral RNA promoters, representing two of the first structures of an RNA-dependent RNA polymerase in complex with its RNA promoters in non-segmented negative-sense RNA viruses. The overall structures of the promoter-bound RSV polymerases are similar to that of the unbound (apo) polymerase. Our structures illustrate the interactions between the RSV polymerase and the RNA promoters and provide the structural basis for the initiation of RNA synthesis at positions 1 and 3 of the RSV promoters. These structures offer a deeper understanding of the pre-initiation state of the RSV polymerase and could aid in antiviral research against RSV.

Structural biology and functional features of phage-derived depolymerase Depo32 on Klebsiella pneumoniae with K2 serotype capsular polysaccharides

Hypervirulent Klebsiella pneumoniae with capsular polysaccharides (CPSs) causes severe nosocomial- and community-acquired infections. Phage-derived depolymerases can degrade CPSs from K. pneumoniae to attenuate bacterial virulence, but their antimicrobial mechanisms and clinical potential are not well understood. In the present study, Klebsiella phage GH-K3-derived depolymerase Depo32 (encoded by gene gp32) was identified to exhibit high efficiency in specifically degrading the CPSs of K2 serotype K. pneumoniae. The cryo-electron microscopy structure of trimeric Depo32 at a resolution up to 2.32 Å revealed potential catalytic centers in the cleft of each of the two adjacent subunits. K. pneumoniae subjected to Depo32 became more sensitive to phagocytosis by RAW264.7 cells and activated the cells by the mitogen-activated protein kinase signaling pathway. In addition, intranasal inoculation with Depo32 (a single dose of 200 µg, 20 µg daily for 3 days, or in combination with gentamicin) rescued all C57BL/6J mice infected with a lethal dose of K. pneumoniae K7 without interference from its neutralizing antibody. In summary, this work elaborates on the mechanism by which Depo32 targets the degradation of K2 serotype CPSs and its potential as an antivirulence agent. IMPORTANCE Depolymerases specific to more than 20 serotypes of Klebsiella spp. have been identified, but most studies only evaluated the single-dose treatment of depolymerases with relatively simple clinical evaluation indices and did not reveal the anti-infection mechanism of these depolymerases in depth. On the basis of determining the biological characteristics, the structure of Depo32 was analyzed by cryo-electron microscopy, and the potential active center was further identified. In addition, the effects of Depo32 on macrophage phagocytosis, signaling pathway activation, and serum killing were revealed, and the efficacy of the depolymerase (single treatment, multiple treatments, or in combination with gentamicin) against acute pneumonia caused by Klebsiella pneumoniae was evaluated. Moreover, the roles of the active sites of Depo32 were also elucidated in the in vitro and in vivo studies. Therefore, through structural biology, cell biology, and in vivo experiments, this study demonstrated the mechanism by which Depo32 targets K2 serotype K. pneumoniae infection.

Novel Artificial Intelligence-Based Approaches for Ab Initio Structure Determination and Atomic Model Building for Cryo-Electron Microscopy

Single particle cryo-electron microscopy (cryo-EM) has emerged as the prevailing method for near-atomic structure determination, shedding light on the important molecular mechanisms of biological macromolecules. However, the inherent dynamics and structural variability of biological complexes coupled with the large number of experimental images generated by a cryo-EM experiment make data processing nontrivial. In particular, ab initio reconstruction and atomic model building remain major bottlenecks that demand substantial computational resources and manual intervention. Approaches utilizing recent innovations in artificial intelligence (AI) technology, particularly deep learning, have the potential to overcome the limitations that cannot be adequately addressed by traditional image processing approaches. Here, we review newly proposed AI-based methods for ab initio volume generation, heterogeneous 3D reconstruction, and atomic model building. We highlight the advancements made by the implementation of AI methods, as well as discuss remaining limitations and areas for future development.

Chirality Analysis of Complex Microparticles using Deep Learning on Realistic Sets of Microscopy Images

Nanoscale chirality is an actively growing research field spurred by the giant chiroptical activity, enantioselective biological activity, and asymmetric catalytic activity of chiral nanostructures. Compared to chiral molecules, the handedness of chiral nano- and microstructures can be directly established via electron microscopy, which can be utilized for the automatic analysis of chiral nanostructures and prediction of their properties. However, chirality in complex materials may have multiple geometric forms and scales. Computational identification of chirality from electron microscopy images rather than optical measurements is convenient but is fundamentally challenging, too, because (1) image features differentiating left- and right-handed particles can be ambiguous and (2) three-dimensional structure essential for chirality is 'flattened' into two-dimensional projections. Here, we show that deep learning algorithms can identify twisted bowtie-shaped microparticles with nearly 100% accuracy and classify them as left- and right-handed with as high as 99% accuracy. Importantly, such accuracy was achieved with as few as 30 original electron microscopy images of bowties. Furthermore, after training on bowtie particles with complex nanostructured features, the model can recognize other chiral shapes with different geometries without retraining for their specific chiral geometry with 93% accuracy, indicating the true learning abilities of the employed neural networks. These findings indicate that our algorithm trained on a practically feasible set of experimental data enables automated analysis of microscopy data for the accelerated discovery of chiral particles and their complex systems for multiple applications.

Multiple-image super-resolution of cryo-electron micrographs based on deep internal learning

Single-particle cryo-electron microscopy (cryo-EM) is a powerful imaging modality capable of visualizing proteins and macromolecular complexes at near-atomic resolution. The low electron-doses used to prevent radiation damage to the biological samples, however, result in images where the power of the noise is 100 times greater than the power of the signal. To overcome these low signal-to-noise ratios (SNRs), hundreds of thousands of particle projections are averaged to determine the three-dimensional structure of the molecule of interest. The sampling requirements of high-resolution imaging impose limitations on the pixel sizes that can be used for acquisition, limiting the size of the field of view and requiring data collection sessions of several days to accumulate sufficient numbers of particles. Meanwhile, recent image super-resolution (SR) techniques based on neural networks have shown state-of-the-art performance on natural images. Building on these advances, here, we present a multiple-image SR algorithm based on deep internal learning designed specifically to work under low-SNR conditions. Our approach leverages the internal image statistics of cryo-EM movies and does not require training on ground-truth data. When applied to single-particle datasets of apoferritin and T20S proteasome, we show that the resolution of the 3D structure obtained from SR micrographs can surpass the limits imposed by the imaging system. Our results indicate that the combination of low magnification imaging with in silico image SR has the potential to accelerate cryo-EM data collection by virtue of including more particles in each exposure and doing so without sacrificing resolution.

Smart de novo Macromolecular Structure Modeling from Cryo-EM Maps

The study of macromolecular structures has expanded our understanding of the amazing cell machinery and such knowledge has changed how the pharmaceutical industry develops new vaccines in recent years. Traditionally, X-ray crystallography has been the main method for structure determination, however, cryogenic electron microscopy (cryo-EM) has increasingly become more popular due to recent advancements in hardware and software. The number of cryo-EM maps deposited in the EMDataResource (formerly EMDatabase) since 2002 has been dramatically increasing and it continues to do so. De novo macromolecular complex modeling is a labor-intensive process, therefore, it is highly desirable to develop software that can automate this process. Here we discuss our automated, data-driven, and artificial intelligence approaches including map processing, feature extraction, modeling building, and target identification. Recently, we have enabled DNA/RNA modeling in our deep learning-based prediction tool, DeepTracer. We have also developed DeepTracer-ID, a tool that can identify proteins solely based on the cryo-EM map. In this paper, we will present our accumulated experiences in developing deep learning-based methods surrounding macromolecule modeling applications.

Emerging Themes in CryoEM─Single Particle Analysis Image Processing

Cryo-electron microscopy (CryoEM) has become a vital technique in structural biology. It is an interdisciplinary field that takes advantage of advances in biochemistry, physics, and image processing, among other disciplines. Innovations in these three basic pillars have contributed to the boosting of CryoEM in the past decade. This work reviews the main contributions in image processing to the current reconstruction workflow of single particle analysis (SPA) by CryoEM. Our review emphasizes the time evolution of the algorithms across the different steps of the workflow differentiating between two groups of approaches: analytical methods and deep learning algorithms. We present an analysis of the current state of the art. Finally, we discuss the emerging problems and challenges still to be addressed in the evolution of CryoEM image processing methods in SPA.

Artificial Intelligence in Cryo-Electron Microscopy

Cryo-electron microscopy (cryo-EM) has become an unrivaled tool for determining the structure of macromolecular complexes. The biological function of macromolecular complexes is inextricably tied to the flexibility of these complexes. Single particle cryo-EM can reveal the conformational heterogeneity of a biochemically pure sample, leading to well-founded mechanistic hypotheses about the roles these complexes play in biology. However, the processing of increasingly large, complex datasets using traditional data processing strategies is exceedingly expensive in both user time and computational resources. Current innovations in data processing capitalize on artificial intelligence (AI) to improve the efficiency of data analysis and validation. Here, we review new tools that use AI to automate the data analysis steps of particle picking, 3D map reconstruction, and local resolution determination. We discuss how the application of AI moves the field forward, and what obstacles remain. We also introduce potential future applications of AI to use cryo-EM in understanding protein communities in cells.

Applications of deep learning in electron microscopy

We review the growing use of machine learning in electron microscopy (EM) driven in part by the availability of fast detectors operating at kiloHertz frame rates leading to large data sets that cannot be processed using manually implemented algorithms. We summarize the various network architectures and error metrics that have been applied to a range of EM-related problems including denoising and inpainting. We then provide a review of the application of these in both physical and life sciences, highlighting how conventional networks and training data have been specifically modified for EM.

Machine Learning for Structure Determination in Single-Particle Cryo-Electron Microscopy: A Systematic Review

Recently, single-particle cryo-electron microscopy (cryo-EM) has become an indispensable method for determining macromolecular structures at high resolution to deeply explore the relevant molecular mechanism. Its recent breakthrough is mainly because of the rapid advances in hardware and image processing algorithms, especially machine learning. As an essential support of single-particle cryo-EM, machine learning has powered many aspects of structure determination and greatly promoted its development. In this article, we provide a systematic review of the applications of machine learning in this field. Our review begins with a brief introduction of single-particle cryo-EM, followed by the specific tasks and challenges of its image processing. Then, focusing on the workflow of structure determination, we describe relevant machine learning algorithms and applications at different steps, including particle picking, 2-D clustering, 3-D reconstruction, and other steps. As different tasks exhibit distinct characteristics, we introduce the evaluation metrics for each task and summarize their dynamics of technology development. Finally, we discuss the open issues and potential trends in this promising field.

Fluorescence sensors for imaging membrane lipid domains and cholesterol

Lipid membrane domains are supramolecular lateral heterogeneities of biological membranes. Of nanoscopic dimensions, they constitute specialized hubs used by the cell as transient signaling platforms for a great variety of biologically important mechanisms. Their property to form and dissolve in the bulk lipid bilayer endow them with the ability to engage in highly dynamic processes, and temporarily recruit subpopulations of membrane proteins in reduced nanometric compartments that can coalesce to form larger mesoscale assemblies. Cholesterol is an essential component of these lipid domains; its unique molecular structure is suitable for interacting intricately with crevices and cavities of transmembrane protein surfaces through its rough β face while "talking" to fatty acid acyl chains of glycerophospholipids and sphingolipids via its smooth α face. Progress in the field of membrane domains has been closely associated with innovative improvements in fluorescence microscopy and new fluorescence sensors. These advances enabled the exploration of the biophysical properties of lipids and their supramolecular platforms. Here I review the rationale behind the use of biosensors over the last few decades and their contributions towards elucidation of the in-plane and transbilayer topography of cholesterol-enriched lipid domains and their molecular constituents. The challenges introduced by super-resolution optical microscopy are discussed, as well as possible scenarios for future developments in the field, including virtual ("no staining") staining.

DeepEMhancer: a deep learning solution for cryo-EM volume post-processing

Cryo-EM maps are valuable sources of information for protein structure modeling. However, due to the loss of contrast at high frequencies, they generally need to be post-processed to improve their interpretability. Most popular approaches, based on global B-factor correction, suffer from limitations. For instance, they ignore the heterogeneity in the map local quality that reconstructions tend to exhibit. Aiming to overcome these problems, we present DeepEMhancer, a deep learning approach designed to perform automatic post-processing of cryo-EM maps. Trained on a dataset of pairs of experimental maps and maps sharpened using their respective atomic models, DeepEMhancer has learned how to post-process experimental maps performing masking-like and sharpening-like operations in a single step. DeepEMhancer was evaluated on a testing set of 20 different experimental maps, showing its ability to reduce noise levels and obtain more detailed versions of the experimental maps. Additionally, we illustrated the benefits of DeepEMhancer on the structure of the SARS-CoV-2 RNA polymerase.

Sanchez-Garcia et al. present DeepEMhancer, a deep learning-based method that can automatically perform post-processing of raw cryo-electron microscopy density maps. The authors report that DeepEMhancer globally improves local quality of density maps, and may represent a useful tool for novel structures where PDB models are not readily available.

Deep learning in electron microscopy

Deep learning is transforming most areas of science and technology, including electron microscopy. This review paper offers a practical perspective aimed at developers with limited familiarity. For context, we review popular applications of deep learning in electron microscopy. Following, we discuss hardware and software needed to get started with deep learning and interface with electron microscopes. We then review neural network components, popular architectures, and their optimization. Finally, we discuss future directions of deep learning in electron microscopy.

Validation tests for cryo-EM maps using an independent particle set

Cryo-electron microscopy (cryo-EM) has revolutionized structural biology by providing 3D density maps of biomolecules at near-atomic resolution. However, map validation is still an open issue. Despite several efforts from the community, it is possible to overfit 3D maps to noisy data. Here, we develop a novel methodology that uses a small independent particle set (not used during the 3D refinement) to validate the maps. The main idea is to monitor how the map probability evolves over the control set during the 3D refinement. The method is complementary to the gold-standard procedure, which generates two reconstructions at each iteration. We low-pass filter the two reconstructions for different frequency cutoffs, and we calculate the probability of each filtered map given the control set. For high-quality maps, the probability should increase as a function of the frequency cutoff and the refinement iteration. We also compute the similarity between the densities of probability distributions of the two reconstructions. As higher frequencies are included, the distributions become more dissimilar. We optimized the BioEM package to perform these calculations, and tested it over systems ranging from quality data to pure noise. Our results show that with our methodology, it possible to discriminate datasets that are constructed from noise particles. We conclude that validation against a control particle set provides a powerful tool to assess the quality of cryo-EM maps.

Advances in methods for atomic resolution macromolecular structure determination

Recent technical advances have dramatically increased the power and scope of structural biology. New developments in high-resolution cryo-electron microscopy, serial X-ray crystallography, and electron diffraction have been especially transformative. Here we highlight some of the latest advances and current challenges at the frontiers of atomic resolution methods for elucidating the structures and dynamical properties of macromolecules and their complexes.

Single-particle cryo-electron microscopy: Mathematical theory, computational challenges, and opportunities

In recent years, an abundance of new molecular structures have been elucidated using cryo-electron microscopy (cryo-EM), largely due to advances in hardware technology and data processing techniques. Owing to these new exciting developments, cryo-EM was selected by Nature Methods as Method of the Year 2015, and the Nobel Prize in Chemistry 2017 was awarded to three pioneers in the field. The main goal of this article is to introduce the challenging and exciting computational tasks involved in reconstructing 3-D molecular structures by cryo-EM. Determining molecular structures requires a wide range of computational tools in a variety of fields, including signal processing, estimation and detection theory, high-dimensional statistics, convex and non-convex optimization, spectral algorithms, dimensionality reduction, and machine learning. The tools from these fields must be adapted to work under exceptionally challenging conditions, including extreme noise levels, the presence of missing data, and massively large datasets as large as several Terabytes. In addition, we present two statistical models: multi-reference alignment and multi-target detection, that abstract away much of the intricacies of cryo-EM, while retaining some of its essential features. Based on these abstractions, we discuss some recent intriguing results in the mathematical theory of cryo-EM, and delineate relations with group theory, invariant theory, and information theory.

Cryo-EM structure of the respiratory syncytial virus RNA polymerase

The respiratory syncytial virus (RSV) RNA polymerase, constituted of a 250 kDa large (L) protein and tetrameric phosphoprotein (P), catalyzes three distinct enzymatic activities — nucleotide polymerization, cap addition, and cap methylation. How RSV L and P coordinate these activities is poorly understood. Here, we present a 3.67 Å cryo-EM structure of the RSV polymerase (L:P) complex. The structure reveals that the RNA dependent RNA polymerase (RdRp) and capping (Cap) domains of L interact with the oligomerization domain (P_OD) and C-terminal domain (P_CTD) of a tetramer of P. The density of the methyltransferase (MT) domain of L and the N-terminal domain of P (P_NTD) is missing. Further analysis and comparison with other RNA polymerases at different stages suggest the structure we obtained is likely to be at an elongation-compatible stage. Together, these data provide enriched insights into the interrelationship, the inhibitors, and the evolutionary implications of the RSV polymerase.

Respiratory syncytial virus (RSV) is a pathogenic non-segmented negative-sense RNA virus and active RSV polymerase is composed of a 250 kDa large (L) protein and tetrameric phosphoprotein (P). Here, the authors present the 3.67 Å cryo-EM structure of the RSV polymerase (L:P) complex.

DeepRes: a new deep-learning- and aspect-based local resolution method for electron-microscopy maps

In this article, a method is presented to estimate a new local quality measure for 3D cryoEM maps that adopts the form of a 'local resolution' type of information. The algorithm (DeepRes) is based on deep-learning 3D feature detection. DeepRes is fully automatic and parameter-free, and avoids the issues of most current methods, such as their insensitivity to enhancements owing to B-factor sharpening (unless the 3D mask is changed), among others, which is an issue that has been virtually neglected in the cryoEM field until now. In this way, DeepRes can be applied to any map, detecting subtle changes in local quality after applying enhancement processes such as isotropic filters or substantially more complex procedures, such as model-based local sharpening, non-model-based methods or denoising, that may be very difficult to follow using current methods. It performs as a human observer expects. The comparison with traditional local resolution indicators is also addressed.