Autofocused 3D Classification of Cryoelectron Subtomograms

Computational methods for in situ structural studies with cryogenic electron tomography

Cryo-electron tomography (cryo-ET) plays a critical role in imaging microorganisms in situ in terms of further analyzing the working mechanisms of viruses and drug exploitation, among others. A data processing workflow for cryo-ET has been developed to reconstruct three-dimensional density maps and further build atomic models from a tilt series of two-dimensional projections. Low signal-to-noise ratio (SNR) and missing wedge are two major factors that make the reconstruction procedure challenging. Because only few near-atomic resolution structures have been reconstructed in cryo-ET, there is still much room to design new approaches to improve universal reconstruction resolutions. This review summarizes classical mathematical models and deep learning methods among general reconstruction steps. Moreover, we also discuss current limitations and prospects. This review can provide software and methods for each step of the entire procedure from tilt series by cryo-ET to 3D atomic structures. In addition, it can also help more experts in various fields comprehend a recent research trend in cryo-ET. Furthermore, we hope that more researchers can collaborate in developing computational methods and mathematical models for high-resolution three-dimensional structures from cryo-ET datasets.

Computational Methods Toward Unbiased Pattern Mining and Structure Determination in Cryo-Electron Tomography Data

Cryo-electron tomography can uniquely probe the native cellular environment for macromolecular structures. Tomograms feature complex data with densities of diverse, densely crowded macromolecular complexes, low signal-to-noise, and artifacts such as the missing wedge effect. Post-processing of this data generally involves isolating regions or particles of interest from tomograms, organizing them into related groups, and rendering final structures through subtomogram averaging. Template-matching and reference-based structure determination are popular analysis methods but are vulnerable to biases and can often require significant user input. Most importantly, these approaches cannot identify novel complexes that reside within the imaged cellular environment. To reliably extract and resolve structures of interest, efficient and unbiased approaches are therefore of great value. This review highlights notable computational software and discusses how they contribute to making automated structural pattern discovery a possibility. Perspectives emphasizing the importance of features for user-friendliness and accessibility are also presented.

High-throughput cryo-ET structural pattern mining by unsupervised deep iterative subtomogram clustering

Cryoelectron tomography directly visualizes heterogeneous macromolecular structures in their native and complex cellular environments. However, existing computer-assisted structure sorting approaches are low throughput or inherently limited due to their dependency on available templates and manual labels. Here, we introduce a high-throughput template-and-label-free deep learning approach, Deep Iterative Subtomogram Clustering Approach (DISCA), that automatically detects subsets of homogeneous structures by learning and modeling 3D structural features and their distributions. Evaluation on five experimental cryo-ET datasets shows that an unsupervised deep learning based method can detect diverse structures with a wide range of molecular sizes. This unsupervised detection paves the way for systematic unbiased recognition of macromolecular complexes in situ.

Towards unsupervised classification of macromolecular complexes in cryo electron tomography: Challenges and opportunities

BACKGROUND AND OBJECTIVES

Cryo electron tomography visualizes native cells at nanometer resolution, but analysis is challenged by noise and artifacts. Recently, supervised deep learning methods have been applied to decipher the 3D spatial distribution of macromolecules. However, in order to discover unknown objects, unsupervised classification techniques are necessary. In this paper, we provide an overview of unsupervised deep learning techniques, discuss the challenges to analyze cryo-ET data, and provide a proof-of-concept on real data.

METHODS

We propose a weakly supervised subtomogram classification method based on transfer learning. We use a deep neural network to learn a clustering friendly representation able to capture 3D shapes in the presence of noise and artifacts. This representation is learned here from a synthetic data set.

RESULTS

We show that when applying k-means clustering given a learning-based representation, it becomes possible to satisfyingly classify real subtomograms according to structural similarity. It is worth noting that no manual annotation is used for performing classification.

CONCLUSIONS

We describe the advantages and limitations of our proof-of-concept and raise several perspectives for improving classification performance.

Subtomogram analysis: The sum of a tomogram's particles reveals molecular structure in situ

Cryo-electron tomography is uniquely suited to provide insights into the molecular architecture of cells and tissue in the native state. While frozen hydrated specimens tolerate sufficient electron doses to distinguish different types of particles in a tomogram, the accumulating beam damage does not allow resolving their detailed molecular structure individually. Statistical methods for subtomogram averaging and classification that coherently enhance the signal of particles corresponding to copies of the same type of macromolecular allow obtaining much higher resolution insights into macromolecules. Here, I review the developments in subtomogram analysis at Wolfgang Baumeister's laboratory that make the dream of structural biology in the native cell become reality.

Cryo-shift: reducing domain shift in cryo-electron subtomograms with unsupervised domain adaptation and randomization

MOTIVATION

Cryo-Electron Tomography (cryo-ET) is a 3D imaging technology that enables the visualization of subcellular structures in situ at near-atomic resolution. Cellular cryo-ET images help in resolving the structures of macromolecules and determining their spatial relationship in a single cell, which has broad significance in cell and structural biology. Subtomogram classification and recognition constitute a primary step in the systematic recovery of these macromolecular structures. Supervised deep learning methods have been proven to be highly accurate and efficient for subtomogram classification, but suffer from limited applicability due to scarcity of annotated data. While generating simulated data for training supervised models is a potential solution, a sizeable difference in the image intensity distribution in generated data as compared with real experimental data will cause the trained models to perform poorly in predicting classes on real subtomograms.

RESULTS

In this work, we present Cryo-Shift, a fully unsupervised domain adaptation and randomization framework for deep learning-based cross-domain subtomogram classification. We use unsupervised multi-adversarial domain adaption to reduce the domain shift between features of simulated and experimental data. We develop a network-driven domain randomization procedure with 'warp' modules to alter the simulated data and help the classifier generalize better on experimental data. We do not use any labeled experimental data to train our model, whereas some of the existing alternative approaches require labeled experimental samples for cross-domain classification. Nevertheless, Cryo-Shift outperforms the existing alternative approaches in cross-domain subtomogram classification in extensive evaluation studies demonstrated herein using both simulated and experimental data.

AVAILABILITYAND IMPLEMENTATION

https://github.com/xulabs/aitom.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

How to build a ribosome from RNA fragments in Chlamydomonas mitochondria

Mitochondria are the powerhouse of eukaryotic cells. They possess their own gene expression machineries where highly divergent and specialized ribosomes, named hereafter mitoribosomes, translate the few essential messenger RNAs still encoded by mitochondrial genomes. Here, we present a biochemical and structural characterization of the mitoribosome in the model green alga Chlamydomonas reinhardtii, as well as a functional study of some of its specific components. Single particle cryo-electron microscopy resolves how the Chlamydomonas mitoribosome is assembled from 13 rRNA fragments encoded by separate non-contiguous gene pieces. Additional proteins, mainly OPR, PPR and mTERF helical repeat proteins, are found in Chlamydomonas mitoribosome, revealing the structure of an OPR protein in complex with its RNA binding partner. Targeted amiRNA silencing indicates that these ribosomal proteins are required for mitoribosome integrity. Finally, we use cryo-electron tomography to show that Chlamydomonas mitoribosomes are attached to the inner mitochondrial membrane via two contact points mediated by Chlamydomonas-specific proteins. Our study expands our understanding of mitoribosome diversity and the various strategies these specialized molecular machines adopt for membrane tethering.

End-to-end robust joint unsupervised image alignment and clustering

Computing dense pixel-to-pixel image correspondences is a fundamental task of computer vision. Often, the objective is to align image pairs from the same semantic category for manipulation or segmentation purposes. Despite achieving superior performance, existing deep learning alignment methods cannot cluster images; consequently, clustering and pairing images needed to be a separate laborious and expensive step. Given a dataset with diverse semantic categories, we propose a multi-task model, Jim-Net, that can directly learn to cluster and align images without any pixel-level or image-level annotations. We design a pair-matching alignment unsupervised training algorithm that selectively matches and aligns image pairs from the clustering branch. Our unsupervised Jim-Net achieves comparable accuracy with state-of-the-art supervised methods on benchmark 2D image alignment dataset PF-PASCAL. Specifically, we apply Jim-Net to cryo-electron tomography, a revolutionary 3D microscopy imaging technique of native subcellular structures. After extensive evaluation on seven datasets, we demonstrate that Jim-Net enables systematic discovery and recovery of representative macromolecular structures in situ, which is essential for revealing molecular mechanisms underlying cellular functions. To our knowledge, Jim-Net is the first end-to-end model that can simultaneously align and cluster images, which significantly improves the performance as compared to performing each task alone.

HEMNMA-3D: Cryo Electron Tomography Method Based on Normal Mode Analysis to Study Continuous Conformational Variability of Macromolecular Complexes

Cryogenic electron tomography (cryo-ET) allows structural determination of biomolecules in their native environment (in situ). Its potential of providing information on the dynamics of macromolecular complexes in cells is still largely unexploited, due to the challenges of the data analysis. The crowded cell environment and continuous conformational changes of complexes make difficult disentangling the data heterogeneity. We present HEMNMA-3D, which is, to the best of our knowledge, the first method for analyzing cryo electron subtomograms in terms of continuous conformational changes of complexes. HEMNMA-3D uses a combination of elastic and rigid-body 3D-to-3D iterative alignments of a flexible 3D reference (atomic structure or electron microscopy density map) to match the conformation, orientation, and position of the complex in each subtomogram. The elastic matching combines molecular mechanics simulation (Normal Mode Analysis of the 3D reference) and experimental, subtomogram data analysis. The rigid-body alignment includes compensation for the missing wedge, due to the limited tilt angle of cryo-ET. The conformational parameters (amplitudes of normal modes) of the complexes in subtomograms obtained through the alignment are processed to visualize the distribution of conformations in a space of lower dimension (typically, 2D or 3D) referred to as space of conformations. This allows a visually interpretable insight into the dynamics of the complexes, by calculating 3D averages of subtomograms with similar conformations from selected (densest) regions and by recording movies of the 3D reference's displacement along selected trajectories through the densest regions. We describe HEMNMA-3D and show its validation using synthetic datasets. We apply HEMNMA-3D to an experimental dataset describing in situ nucleosome conformational variability. HEMNMA-3D software is available freely (open-source) as part of ContinuousFlex plugin of Scipion V3.0 (http://scipion.i2pc.es).

Molecular and cellular dynamics of the 26S proteasome

In eukaryotic cells, the ubiquitin-proteasome system serves to remove proteins that are either dysfunctional or no longer needed. The 26S proteasome is a 2.5 MDa multisubunit complex comprising the 20S core particle, where degradation is executed, and one or two regulatory particles which prepare substrates for degradation. Whereas the 20S core particles of several species had been studied extensively by X-ray crystallography, the 26S holocomplex structure had remained elusive for a long time. Recent advances in single-particle cryo-electron microscopy have changed the situation and provided atomic resolution models of this intriguing molecular machine and its dynamics. Besides, cryo-electron tomography enables structural studies in situ, providing molecular resolution images of macromolecules inside pristinely preserved cellular environments. This has greatly contributed to our understanding of proteasome dynamics in the context of cells.

The promise and the challenges of cryo-electron tomography

Structural biologists have traditionally approached cellular complexity in a reductionist manner in which the cellular molecular components are fractionated and purified before being studied individually. This 'divide and conquer' approach has been highly successful. However, awareness has grown in recent years that biological functions can rarely be attributed to individual macromolecules. Most cellular functions arise from their concerted action, and there is thus a need for methods enabling structural studies performed in situ, ideally in unperturbed cellular environments. Cryo-electron tomography (Cryo-ET) combines the power of 3D molecular-level imaging with the best structural preservation that is physically possible to achieve. Thus, it has a unique potential to reveal the supramolecular architecture or 'molecular sociology' of cells and to discover the unexpected. Here, we review state-of-the-art Cryo-ET workflows, provide examples of biological applications, and discuss what is needed to realize the full potential of Cryo-ET.

Spark-based parallel calculation of 3D fourier shell correlation for macromolecule structure local resolution estimation

Background

Resolution estimation is the main evaluation criteria for the reconstruction of macromolecular 3D structure in the field of cryoelectron microscopy (cryo-EM). At present, there are many methods to evaluate the 3D resolution for reconstructed macromolecular structures from Single Particle Analysis (SPA) in cryo-EM and subtomogram averaging (SA) in electron cryotomography (cryo-ET). As global methods, they measure the resolution of the structure as a whole, but they are inaccurate in detecting subtle local changes of reconstruction. In order to detect the subtle changes of reconstruction of SPA and SA, a few local resolution methods are proposed. The mainstream local resolution evaluation methods are based on local Fourier shell correlation (FSC), which is computationally intensive. However, the existing resolution evaluation methods are based on multi-threading implementation on a single computer with very poor scalability.

Results

This paper proposes a new fine-grained 3D array partition method by key-value format in Spark. Our method first converts 3D images to key-value data (K-V). Then the K-V data is used for 3D array partitioning and data exchange in parallel. So Spark-based distributed parallel computing framework can solve the above scalability problem. In this distributed computing framework, all 3D local FSC tasks are simultaneously calculated across multiple nodes in a computer cluster. Through the calculation of experimental data, 3D local resolution evaluation algorithm based on Spark fine-grained 3D array partition has a magnitude change in computing speed compared with the mainstream FSC algorithm under the condition that the accuracy remains unchanged, and has better fault tolerance and scalability.

Conclusions

In this paper, we proposed a K-V format based fine-grained 3D array partition method in Spark to parallel calculating 3D FSC for getting a 3D local resolution density map. 3D local resolution density map evaluates the three-dimensional density maps reconstructed from single particle analysis and subtomogram averaging. Our proposed method can significantly increase the speed of the 3D local resolution evaluation, which is important for the efficient detection of subtle variations among reconstructed macromolecular structures.

A clearer picture of the ER translocon complex

The endoplasmic reticulum (ER) translocon complex is the main gate into the secretory pathway, facilitating the translocation of nascent peptides into the ER lumen or their integration into the lipid membrane. Protein biogenesis in the ER involves additional processes, many of them occurring co-translationally while the nascent protein resides at the translocon complex, including recruitment of ER-targeted ribosome-nascent-chain complexes, glycosylation, signal peptide cleavage, membrane protein topogenesis and folding. To perform such varied functions on a broad range of substrates, the ER translocon complex has different accessory components that associate with it either stably or transiently. Here, we review recent structural and functional insights into this dynamically constituted central hub in the ER and its components. Recent cryo-electron microscopy (EM) studies have dissected the molecular organization of the co-translational ER translocon complex, comprising the Sec61 protein-conducting channel, the translocon-associated protein complex and the oligosaccharyl transferase complex. Complemented by structural characterization of the post-translational import machinery, key molecular principles emerge that distinguish co- and post-translational protein import and biogenesis. Further cryo-EM structures promise to expand our mechanistic understanding of the various biochemical functions involving protein biogenesis and quality control in the ER.

Fine-grained alignment of cryo-electron subtomograms based on MPI parallel optimization

BACKGROUND

Cryo-electron tomography (Cryo-ET) is an imaging technique used to generate three-dimensional structures of cellular macromolecule complexes in their native environment. Due to developing cryo-electron microscopy technology, the image quality of three-dimensional reconstruction of cryo-electron tomography has greatly improved. However, cryo-ET images are characterized by low resolution, partial data loss and low signal-to-noise ratio (SNR). In order to tackle these challenges and improve resolution, a large number of subtomograms containing the same structure needs to be aligned and averaged. Existing methods for refining and aligning subtomograms are still highly time-consuming, requiring many computationally intensive processing steps (i.e. the rotations and translations of subtomograms in three-dimensional space).

RESULTS

In this article, we propose a Stochastic Average Gradient (SAG) fine-grained alignment method for optimizing the sum of dissimilarity measure in real space. We introduce a Message Passing Interface (MPI) parallel programming model in order to explore further speedup.

CONCLUSIONS

We compare our stochastic average gradient fine-grained alignment algorithm with two baseline methods, high-precision alignment and fast alignment. Our SAG fine-grained alignment algorithm is much faster than the two baseline methods. Results on simulated data of GroEL from the Protein Data Bank (PDB ID:1KP8) showed that our parallel SAG-based fine-grained alignment method could achieve close-to-optimal rigid transformations with higher precision than both high-precision alignment and fast alignment at a low SNR (SNR=0.003) with tilt angle range ±60∘ or ±40∘. For the experimental subtomograms data structures of GroEL and GroEL/GroES complexes, our parallel SAG-based fine-grained alignment can achieve higher precision and fewer iterations to converge than the two baseline methods.

An integration of fast alignment and maximum-likelihood methods for electron subtomogram averaging and classification

Motivation

Cellular Electron CryoTomography (CECT) is an emerging 3D imaging technique that visualizes subcellular organization of single cells at sub-molecular resolution and in near-native state. CECT captures large numbers of macromolecular complexes of highly diverse structures and abundances. However, the structural complexity and imaging limits complicate the systematic de novo structural recovery and recognition of these macromolecular complexes. Efficient and accurate reference-free subtomogram averaging and classification represent the most critical tasks for such analysis. Existing subtomogram alignment based methods are prone to the missing wedge effects and low signal-to-noise ratio (SNR). Moreover, existing maximum-likelihood based methods rely on integration operations, which are in principle computationally infeasible for accurate calculation.

Results

Built on existing works, we propose an integrated method, Fast Alignment Maximum Likelihood method (FAML), which uses fast subtomogram alignment to sample sub-optimal rigid transformations. The transformations are then used to approximate integrals for maximum-likelihood update of subtomogram averages through expectation–maximization algorithm. Our tests on simulated and experimental subtomograms showed that, compared to our previously developed fast alignment method (FA), FAML is significantly more robust to noise and missing wedge effects with moderate increases of computation cost. Besides, FAML performs well with significantly fewer input subtomograms when the FA method fails. Therefore, FAML can serve as a key component for improved construction of initial structural models from macromolecules captured by CECT.

Availability and implementation

http://www.cs.cmu.edu/mxu1

Adversarial domain adaptation for cross data source macromolecule in situ structural classification in cellular electron cryo-tomograms

MOTIVATION

Since 2017, an increasing amount of attention has been paid to the supervised deep learning-based macromolecule in situ structural classification (i.e. subtomogram classification) in cellular electron cryo-tomography (CECT) due to the substantially higher scalability of deep learning. However, the success of such supervised approach relies heavily on the availability of large amounts of labeled training data. For CECT, creating valid training data from the same data source as prediction data is usually laborious and computationally intensive. It would be beneficial to have training data from a separate data source where the annotation is readily available or can be performed in a high-throughput fashion. However, the cross data source prediction is often biased due to the different image intensity distributions (a.k.a. domain shift).

RESULTS

We adapt a deep learning-based adversarial domain adaptation (3D-ADA) method to timely address the domain shift problem in CECT data analysis. 3D-ADA first uses a source domain feature extractor to extract discriminative features from the training data as the input to a classifier. Then it adversarially trains a target domain feature extractor to reduce the distribution differences of the extracted features between training and prediction data. As a result, the same classifier can be directly applied to the prediction data. We tested 3D-ADA on both experimental and realistically simulated subtomogram datasets under different imaging conditions. 3D-ADA stably improved the cross data source prediction, as well as outperformed two popular domain adaptation methods. Furthermore, we demonstrate that 3D-ADA can improve cross data source recovery of novel macromolecular structures.

AVAILABILITY AND IMPLEMENTATION

https://github.com/xulabs/projects.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

In situ structure determination by subtomogram averaging

Cryo-tomography and subtomogram averaging are increasingly popular techniques for structural determination of macromolecular complexes in situ. They have the potential to achieve high-resolution views of native complexes, together with the details of their location relative to interacting molecules. The subtomogram averaging (StA) pipelines are well-established, with current developments aiming to optimise each step by reducing manual intervention and user decisions, following similar trends in single-particle approaches that have dramatically increased their popularity. Here, we review the main steps of typical StA workflows. We focus on considerations arising from the fact that the objects of study are embedded within unique crowded environments, and we emphasise those steps where careful decisions need to be made by the user.

Subtomogram averaging from cryo-electron tomograms

Cryo-electron tomography (cryo-ET) allows three-dimensional (3D) visualization of frozen-hydrated biological samples, such as protein complexes and cell organelles, in near-native environments at nanometer scale. Protein complexes that are present in multiple copies in a set of tomograms can be extracted, mutually aligned, and averaged to yield a signal-enhanced 3D structure up to sub-nanometer or even near-atomic resolution. This technique, called subtomogram averaging (StA), is powered by improvements in EM hardware and image processing software. Importantly, StA provides unique biological insights into the structure and function of cellular machinery in close-to-native contexts. In this chapter, we describe the principles and key steps of StA. We briefly cover sample preparation and data collection with an emphasis on image processing procedures related to tomographic reconstruction, subtomogram alignment, averaging, and classification. We conclude by summarizing current limitations and future directions of this technique with a focus on high-resolution StA.

Toward high-resolution in situ structural biology with cryo-electron tomography and subtomogram averaging

Cryo-electron tomography (cryo-ET) provides unprecedented insights into the molecular constituents of biological environments. In combination with an image processing method called subtomogram averaging (STA), detailed 3D structures of biological molecules can be obtained in large, irregular macromolecular assemblies or in situ, without the need for purification. The contextual meta-information these methods also provide, such as a protein's location within its native environment, can then be combined with functional data. This allows the derivation of a detailed view on the physiological or pathological roles of proteins from the molecular to cellular level. Despite their tremendous potential in in situ structural biology, cryo-ET and STA have been restricted by methodological limitations, such as the low obtainable resolution. Exciting progress now allows one to reach unprecedented resolutions in situ, ranging in optimal cases beyond the nanometer barrier. Here, I review current frontiers and future challenges in routinely determining high-resolution structures in in situ environments using cryo-ET and STA.

Unravelling molecular complexity in structural cell biology

Structural and cell biology have traditionally been separate disciplines and employed techniques that were well defined within the realm of either one or the other. Recent technological breakthroughs propelled electron microscopy of frozen hydrated specimens (cryo-EM) followed by single-particle analysis (SPA) to become a widely applied approach for obtaining near-atomic resolution structures of purified macromolecules. In parallel, ongoing developments on sample preparation are increasingly successful in bringing molecular views into cell biology. Cryo-electron tomography (cryo-ET) has so far served as the main imaging modality employed in these efforts towards obtaining three-dimensional (3D) volumes of heterogeneous molecular assemblies. We review the state-of-the-art in cryo-ET and computational processing and describe the current opportunities and frontiers for in-cell applications.

DEEP LEARNING BASED SUPERVISED SEMANTIC SEGMENTATION OF ELECTRON CRYO-SUBTOMOGRAMS

Cellular Electron Cryo-Tomography (CECT) is a powerful imaging technique for the 3D visualization of cellular structure and organization at submolecular resolution. It enables analyzing the native structures of macromolecular complexes and their spatial organization inside single cells. However, due to the high degree of structural complexity and practical imaging limitations, systematic macromolecular structural recovery inside CECT images remains challenging. Particularly, the recovery of a macromolecule is likely to be biased by its neighbor structures due to the high molecular crowding. To reduce the bias, here we introduce a novel 3D convolutional neural network inspired by Fully Convolutional Network and Encoder-Decoder Architecture for the supervised segmentation of macromolecules of interest in subtomograms. The tests of our models on realistically simulated CECT data demonstrate that our new approach has significantly improved segmentation performance compared to our baseline approach. Also, we demonstrate that the proposed model has generalization ability to segment new structures that do not exist in training data.

Cryo-electron tomography for the structural study of mitochondrial translation

Cryo-electron tomography (cryo-ET) enables the three-dimensional (3D) structural characterization of macromolecular complexes in their physiological environment. Thus, cryo-ET is uniquely suited to study the structural basis of biomolecular processes that are extremely difficult or even impossible to reconstitute using purified components. Translation of mitochondrial genes, which occurs in the secluded interior of mitochondria, falls into this category. Here, we describe the principles of cryo-ET in the context of mitochondrial translation and outline recent developments and challenges of the method. The 3D image of a frozen-hydrated biological sample is computed from its 2D projections, which are acquired using a transmission electron microscope. In conjunction with automated detection of different copies of the molecule of interest and averaging of the corresponding subtomograms, cryo-ET enables macromolecular structure determination in the native environment (i.e. in situ) at sub-nanometer resolution. The preservation of the native environment furthermore allows the extraction of contextual information about the molecules, including the location of specific molecules with respect to membranes, their relative positioning and the spatial organization with respect to other types of macromolecules. Recent preparative developments extend the field of application of cryo-ET from isolated organelles to cultured eukaryotic cells and even tissue, making the traditional borders between molecular and cellular structural biology disappear.

A convolutional autoencoder approach for mining features in cellular electron cryo-tomograms and weakly supervised coarse segmentation

Cellular electron cryo-tomography enables the 3D visualization of cellular organization in the near-native state and at submolecular resolution. However, the contents of cellular tomograms are often complex, making it difficult to automatically isolate different in situ cellular components. In this paper, we propose a convolutional autoencoder-based unsupervised approach to provide a coarse grouping of 3D small subvolumes extracted from tomograms. We demonstrate that the autoencoder can be used for efficient and coarse characterization of features of macromolecular complexes and surfaces, such as membranes. In addition, the autoencoder can be used to detect non-cellular features related to sample preparation and data collection, such as carbon edges from the grid and tomogram boundaries. The autoencoder is also able to detect patterns that may indicate spatial interactions between cellular components. Furthermore, we demonstrate that our autoencoder can be used for weakly supervised semantic segmentation of cellular components, requiring a very small amount of manual annotation.

Deep learning-based subdivision approach for large scale macromolecules structure recovery from electron cryo tomograms

Motivation

Cellular Electron CryoTomography (CECT) enables 3D visualization of cellular organization at near-native state and in sub-molecular resolution, making it a powerful tool for analyzing structures of macromolecular complexes and their spatial organizations inside single cells. However, high degree of structural complexity together with practical imaging limitations makes the systematic de novo discovery of structures within cells challenging. It would likely require averaging and classifying millions of subtomograms potentially containing hundreds of highly heterogeneous structural classes. Although it is no longer difficult to acquire CECT data containing such amount of subtomograms due to advances in data acquisition automation, existing computational approaches have very limited scalability or discrimination ability, making them incapable of processing such amount of data.

Results

To complement existing approaches, in this article we propose a new approach for subdividing subtomograms into smaller but relatively homogeneous subsets. The structures in these subsets can then be separately recovered using existing computation intensive methods. Our approach is based on supervised structural feature extraction using deep learning, in combination with unsupervised clustering and reference-free classification. Our experiments show that, compared with existing unsupervised rotation invariant feature and pose-normalization based approaches, our new approach achieves significant improvements in both discrimination ability and scalability. More importantly, our new approach is able to discover new structural classes and recover structures that do not exist in training data.

Availability and Implementation

Source code freely available at http://www.cs.cmu.edu/∼mxu1/software.

Supplementary information

Supplementary data are available at Bioinformatics online.

Proteasomes tether to two distinct sites at the nuclear pore complex

This study compares the native structures of cytosolic and nuclear proteasomes, visualized directly within cells. The assembly states and functional states of proteasomes in each compartment were similar, indicating comparable levels of proteolytic activity per proteasome. Nuclear proteasomes were tethered to two different sites at the nuclear pore complex (NPC): the inner nuclear membrane and the NPC basket. Structural analysis revealed mechanistic details of the two tethering interactions. These results present direct evidence that proteasomes bind at NPCs, establishing a cellular hub for protein degradation at the gateway between the nucleus and cytoplasm. This work demonstrates how cryo-electron tomography can reveal biological mechanisms by directly observing the interactions between molecular complexes within the native cellular environment.

The partitioning of cellular components between the nucleus and cytoplasm is the defining feature of eukaryotic life. The nuclear pore complex (NPC) selectively gates the transport of macromolecules between these compartments, but it is unknown whether surveillance mechanisms exist to reinforce this function. By leveraging in situ cryo-electron tomography to image the native cellular environment of Chlamydomonas reinhardtii, we observed that nuclear 26S proteasomes crowd around NPCs. Through a combination of subtomogram averaging and nanometer-precision localization, we identified two classes of proteasomes tethered via their Rpn9 subunits to two specific NPC locations: binding sites on the NPC basket that reflect its eightfold symmetry and more abundant binding sites at the inner nuclear membrane that encircle the NPC. These basket-tethered and membrane-tethered proteasomes, which have similar substrate-processing state frequencies as proteasomes elsewhere in the cell, are ideally positioned to regulate transcription and perform quality control of both soluble and membrane proteins transiting the NPC.

TomoMiner and TomoMinerCloud: A Software Platform for Large-Scale Subtomogram Structural Analysis

Cryo-electron tomography (cryo-ET) captures the 3D electron density distribution of macromolecular complexes in close to native state. With the rapid advance of cryo-ET acquisition technologies, it is possible to generate large numbers (>100,000) of subtomograms, each containing a macromolecular complex. Often, these subtomograms represent a heterogeneous sample due to variations in the structure and composition of a complex in situ form or because particles are a mixture of different complexes. In this case subtomograms must be classified. However, classification of large numbers of subtomograms is a time-intensive task and often a limiting bottleneck. This paper introduces an open source software platform, TomoMiner, for large-scale subtomogram classification, template matching, subtomogram averaging, and alignment. Its scalable and robust parallel processing allows efficient classification of tens to hundreds of thousands of subtomograms. In addition, TomoMiner provides a pre-configured TomoMinerCloud computing service permitting users without sufficient computing resources instant access to TomoMiners high-performance features.

In situ structural studies of tripeptidyl peptidase II (TPPII) reveal spatial association with proteasomes

Tripeptidyl peptidase II (TPPII) is a eukaryotic protease acting downstream of the 26S proteasome; it removes tripeptides from the degradation products released by the proteasome. Structural studies in vitro have revealed the basic architecture of TPPII, a two-stranded linear polymer that assembles to form a spindle-shaped complex of ∼6 MDa. Dependent on protein concentration, TPPII has a distinct tendency for polymorphism. Therefore, its structure in vivo has remained unclear. To resolve this issue, we have scrutinized cryo-electron tomograms of rat hippocampal neurons for the occurrence and spatial distribution of TPPII by template matching. The quality of the tomograms recorded with the Volta phase plate enabled a detailed structural analysis of TPPII despite its low abundance. Two different assembly states (36-mers and 32-mers) coexist as well as occasional extended forms with longer strands. A distance analysis of the relative locations of TPPII and 26S proteasomes confirmed the visual impression that these two complexes spatially associate in agreement with TPPII's role in postproteasomal degradation.

The integrative role of cryo electron microscopy in molecular and cellular structural biology

After gradually moving away from preparation methods prone to artefacts such as plastic embedding and negative staining for cell sections and single particles, the field of cryo electron microscopy (cryo-EM) is now heading off at unprecedented speed towards high-resolution analysis of biological objects of various sizes. This 'revolution in resolution' is happening largely thanks to new developments of new-generation cameras used for recording the images in the cryo electron microscope which have much increased sensitivity being based on complementary metal oxide semiconductor devices. Combined with advanced image processing and 3D reconstruction, the cryo-EM analysis of nucleoprotein complexes can provide unprecedented insights at molecular and atomic levels and address regulatory mechanisms in the cell. These advances reinforce the integrative role of cryo-EM in synergy with other methods such as X-ray crystallography, fluorescence imaging or focussed-ion beam milling as exemplified here by some recent studies from our laboratory on ribosomes, viruses, chromatin and nuclear receptors. Such multi-scale and multi-resolution approaches allow integrating molecular and cellular levels when applied to purified or in situ macromolecular complexes, thus illustrating the trend of the field towards cellular structural biology.

Cryo-Electron Tomography and Subtomogram Averaging

Cryo-electron tomography (cryo-ET) allows 3D volumes to be reconstructed from a set of 2D projection images of a tilted biological sample. It allows densities to be resolved in 3D that would otherwise overlap in 2D projection images. Cryo-ET can be applied to resolve structural features in complex native environments, such as within the cell. Analogous to single-particle reconstruction in cryo-electron microscopy, structures present in multiple copies within tomograms can be extracted, aligned, and averaged, thus increasing the signal-to-noise ratio and resolution. This reconstruction approach, termed subtomogram averaging, can be used to determine protein structures in situ. It can also be applied to facilitate more conventional 2D image analysis approaches. In this chapter, we provide an introduction to cryo-ET and subtomogram averaging. We describe the overall workflow, including tomographic data collection, preprocessing, tomogram reconstruction, subtomogram alignment and averaging, classification, and postprocessing. We consider theoretical issues and practical considerations for each step in the workflow, along with descriptions of recent methodological advances and remaining limitations.

Cryo-EM and the elucidation of new macromolecular structures: Random Conical Tilt revisited

Cryo-Electron Microscopy (cryo-EM) of macromolecular complexes is a fundamental structural biology technique which is expanding at a very fast pace. Key to its success in elucidating the three-dimensional structure of a macromolecular complex, especially of small and non-symmetric ones, is the ability to start from a low resolution map, which is subsequently refined with the actual images collected at the microscope. There are several methods to produce this first structure. Among them, Random Conical Tilt (RCT) plays a prominent role due to its unbiased nature (it can create an initial model based on experimental measurements). In this article, we revise the fundamental mathematical expressions supporting RCT, providing new expressions handling all key geometrical parameters without the need of intermediate operations, leading to improved automation and overall reliability, essential for the success of cryo-EM when analyzing new complexes. We show that the here proposed RCT workflow based on the new formulation performs very well in practical cases, requiring very few image pairs (as low as 13 image pairs in one of our examples) to obtain relevant 3D maps.

Proteasomes. A molecular census of 26S proteasomes in intact neurons

The 26S proteasome is a key player in eukaryotic protein quality control and in the regulation of numerous cellular processes. Here, we describe quantitative in situ structural studies of this highly dynamic molecular machine in intact hippocampal neurons. We used electron cryotomography with the Volta phase plate, which allowed high fidelity and nanometer precision localization of 26S proteasomes. We undertook a molecular census of single- and double-capped proteasomes and assessed the conformational states of individual complexes. Under the conditions of the experiment—that is, in the absence of proteotoxic stress—only 20% of the 26S proteasomes were engaged in substrate processing. The remainder was in the substrate-accepting ground state. These findings suggest that in the absence of stress, the capacity of the proteasome system is not fully used.