TOMOMAN: a software package for large scale cryo-electron tomography data preprocessing, community data sharing, and collaborative computing.

Cryo-electron tomography (cryo-ET) and subtomogram averaging (STA) are becoming the preferred methodologies for investigating subcellular and macromolecular structures in native or near-native environments. While cryo-ET is amenable to a wide range of biological problems, these problems often have data processing requirements that need to be individually optimized, precluding the notion of a one-size-fits-all processing pipeline. Cryo-ET data processing is also becoming progressively more complex due to an increasing number of packages for each processing step. Though each package has its own strengths and weaknesses, independent development and different data formats makes them difficult to interface with one another. TOMOMAN (TOMOgram MANager) is an extensible package for streamlining the interoperability of packages, enabling users to develop project-specific processing workflows. TOMOMAN does this by maintaining an internal metadata format and wrapping external packages to manage and perform preprocessing, from raw tilt-series data to reconstructed tomograms. TOMOMAN can also export this metadata between various STA packages. TOMOMAN also includes tools for archiving projects to data repositories; allowing subsequent users to download TOMOMAN projects and directly resume processing where it was previously left off. By tracking essential metadata, TOMOMAN streamlines data sharing, which improves reproducibility of published results, reduces computational costs by minimizing reprocessing, and enables distributed cryo-ET projects between multiple groups and institutions. TOMOMAN provides a way for users to test different software packages to develop processing workflows that meet the specific needs of their biological questions and to distribute their results with the broader scientific community.

STOPGAP: an open-source package for template matching, subtomogram alignment and classification.

Cryo-electron tomography (cryo-ET) enables molecular-resolution 3D imaging of complex biological specimens such as viral particles, cellular sections and, in some cases, whole cells. This enables the structural characterization of molecules in their near-native environments, without the need for purification or separation, thereby preserving biological information such as conformational states and spatial relationships between different molecular species. Subtomogram averaging is an image-processing workflow that allows users to leverage cryo-ET data to identify and localize target molecules, determine high-resolution structures of repeating molecular species and classify different conformational states. Here, STOPGAP, an open-source package for subtomogram averaging that is designed to provide users with fine control over each of these steps, is described. In providing detailed descriptions of the image-processing algorithms that STOPGAP uses, this manuscript is also intended to serve as a technical resource to users as well as for further community-driven software development.

STOPGAP, an open-source package for template matching, subtomogram alignment, and classification.

Cryo-electron tomography (cryo-ET) enables molecular-resolution 3D imaging of complex biological specimens such as viral particles, cellular sections, and in some cases, whole cells. This enables the structural characterization of molecules in their near-native environments, without the need for purification or separation, thereby preserving biological information such as conformational states and spatial relationships between different molecular species. Subtomogram averaging is an image processing workflow that allows users to leverage cryo-ET data to identify and localize target molecules, determine high-resolution structures of repeating molecular species, and classifying different conformational states. Here we describe STOPGAP, an open-source package for subtomogram averaging designed to provide users with fine control over each of these steps. In providing detailed descriptions of the image processing algorithms that STOPGAP uses, we intend for this manuscript to also serve as a technical resource to users as well as further community-driven software development.

Volumetric macromolecule identification in cryo-electron tomograms using capsule networks.

BACKGROUND: Despite recent advances in cellular cryo-electron tomography (CET), developing automated tools for macromolecule identification in submolecular resolution remains challenging due to the lack of annotated data and high structural complexities. To date, the extent of the deep learning methods constructed for this problem is limited to conventional Convolutional Neural Networks (CNNs). Identifying macromolecules of different types and sizes is a tedious and time-consuming task. In this paper, we employ a capsule-based architecture to automate the task of macromolecule identification, that we refer to as 3D-UCaps. In particular, the architecture is composed of three components: feature extractor, capsule encoder, and CNN decoder. The feature extractor converts voxel intensities of input sub-tomograms to activities of local features. The encoder is a 3D Capsule Network (CapsNet) that takes local features to generate a low-dimensional representation of the input. Then, a 3D CNN decoder reconstructs the sub-tomograms from the given representation by upsampling. RESULTS: We performed binary and multi-class localization and identification tasks on synthetic and experimental data. We observed that the 3D-UNet and the 3D-UCaps had an [Formula: see text]score mostly above 60% and 70%, respectively, on the test data. In both network architectures, we observed degradation of at least 40% in the [Formula: see text]-score when identifying very small particles (PDB entry 3GL1) compared to a large particle (PDB entry 4D8Q). In the multi-class identification task of experimental data, 3D-UCaps had an [Formula: see text]-score of 91% on the test data in contrast to 64% of the 3D-UNet. The better [Formula: see text]-score of 3D-UCaps compared to 3D-UNet is obtained by a higher precision score. We speculate this to be due to the capsule network employed in the encoder. To study the effect of the CapsNet-based encoder architecture further, we performed an ablation study and perceived that the [Formula: see text]-score is boosted as network depth is increased which is in contrast to the previously reported results for the 3D-UNet. To present a reproducible work, source code, trained models, data as well as visualization results are made publicly available. CONCLUSION: Quantitative and qualitative results show that 3D-UCaps successfully perform various downstream tasks including identification and localization of macromolecules and can at least compete with CNN architectures for this task. Given that the capsule layers extract both the existence probability and the orientation of the molecules, this architecture has the potential to lead to representations of the data that are better interpretable than those of 3D-UNet.

Routine Collection of High-Resolution cryo-EM Datasets Using 200 KV Transmission Electron Microscope.

Cryo-electron microscopy (cryo-EM) has been established as a routine method for protein structure determination during the past decade, taking an ever-increasing share of published structural data. Recent advances in TEM technology and automation have boosted both the speed of data collection and quality of acquired images while simultaneously decreasing the required level of expertise for obtaining cryo-EM maps at sub-3 Å resolutions. While most of such high-resolution structures have been obtained using state-of-the-art 300 kV cryo-TEM systems, high-resolution structures can be also obtained with 200 kV cryo-TEM systems, especially when equipped with an energy filter. Additionally, automation of microscope alignments and data collection with real-time image quality assessment reduces system complexity and assures optimal microscope settings, resulting in increased yield of high-quality images and overall throughput of data collection. This protocol demonstrates the implementation of recent technological advances and automation features on a 200 kV cryo-transmission electron microscope and shows how to collect data for the reconstruction of 3D maps that are sufficient for de novo atomic model building. We focus on best practices, critical variables, and common issues that must be considered to enable the routine collection of such high-resolution cryo-EM datasets. Particularly the following essential topics are reviewed in detail: i) automation of microscope alignments, ii) selection of suitable areas for data acquisition, iii) optimal optical parameters for high-quality, high-throughput data collection, iv) energy filter tuning for zero-loss imaging, and v) data management and quality assessment. Application of the best practices and improvement of achievable resolution using an energy filter will be demonstrated on the example of apo-ferritin that was reconstructed to 1.6 Å, and Thermoplasma acidophilum 20S proteasome reconstructed to 2.1-Å resolution using a 200 kV TEM equipped with an energy filter and a direct electron detector.

In situ cryo-electron tomography reveals gradient organization of ribosome biogenesis in intact nucleoli.

Ribosomes comprise a large (LSU) and a small subunit (SSU) which are synthesized independently in the nucleolus before being exported into the cytoplasm, where they assemble into functional ribosomes. Individual maturation steps have been analyzed in detail using biochemical methods, light microscopy and conventional electron microscopy (EM). In recent years, single particle analysis (SPA) has yielded molecular resolution structures of several pre-ribosomal intermediates. It falls short, however, of revealing the spatiotemporal sequence of ribosome biogenesis in the cellular context. Here, we present our study on native nucleoli in Chlamydomonas reinhardtii, in which we follow the formation of LSU and SSU precursors by in situ cryo-electron tomography (cryo-ET) and subtomogram averaging (STA). By combining both positional and molecular data, we reveal gradients of ribosome maturation within the granular component (GC), offering a new perspective on how the liquid-liquid-phase separation of the nucleolus supports ribosome biogenesis.

Conformational dynamics of Peb4 exhibit "mother's arms" chain model: a molecular dynamics study.

Peb4 from Campylobacter jejuni is an intertwined dimeric, periplasmic holdase, which also exhibits peptidyl prolyl cis/trans isomerase (PPIase) activity. Peb4 gene deletion alters the outer membrane protein profile and impairs cellular adhesion and biofilm formation for C. jejuni. Earlier crystallographic study has proposed that the PPIase domains are flexible and might form a cradle for holding the substrate and these aspects of Peb4 were explored using sub-microsecond molecular dynamics simulations in solution environment. Our simulations have revealed that PPIase domains are highly flexible and undergo a large structural change where they move apart from each other by 8 nm starting at .5 nm. Further, this large conformational change renders Peb4 as a compact protein with crossed-over conformation, forms a central cavity, which can "cradle" the target substrate. As reported for other chaperone proteins, flexibility of linker region connecting the chaperone and PPIase domains is key to forming the "crossed-over" conformation. The conformational transition of the Peb4 protein from the X-ray structure to the crossed-over conformation follows the "mother's arms" chain model proposed for the FkpA chaperone protein. Our results offer insights into how Peb4 and similar chaperones can use the conformational heterogeneity at their disposal to perform its much-revered biological function.

Structural insights into dynamics of RecU-HJ complex formation elucidates key role of NTR and stalk region toward formation of reactive state.

Holliday junction (HJ) resolving enzyme RecU is involved in DNA repair and recombination. We have determined the crystal structure of inactive mutant (D88N) of RecU from Bacillus subtilis in complex with a 12 base palindromic DNA fragment at a resolution of 3.2 Å. This structure shows the stalk region and the essential N-terminal region (NTR) previously unseen in our DNA unbound structure. The flexible nature of the NTR in solution was confirmed using SAXS. Thermofluor studies performed to assess the stability of RecU in complex with the arms of an HJ indicate that it confers stability. Further, we performed molecular dynamics (MD) simulations of wild type and an NTR deletion variant of RecU, with and without HJ. The NTR is observed to be highly flexible in simulations of the unbound RecU, in agreement with SAXS observations. These simulations revealed domain dynamics of RecU and their role in the formation of complex with HJ. The MD simulations also elucidate key roles of the NTR, stalk region, and breathing motion of RecU in the formation of the reactive state.

Time-dependent X-ray diffraction studies on urea/hen egg white lysozyme complexes reveal structural changes that indicate onset of denaturation.

Temporal binding of urea to lysozyme was examined using X-ray diffraction of single crystals of urea/lysozyme complexes prepared by soaking native lysozyme crystals in solutions containing 9 M urea. Four different soak times of 2, 4, 7 and 10 hours were used. The five crystal structures (including the native lysozyme), refined to 1.6 Å resolution, reveal that as the soaking time increased, more and more first-shell water molecules are replaced by urea. The number of hydrogen bonds between urea and the protein is similar to that between protein and water molecules replaced by urea. However, the number of van der Waals contacts to protein from urea is almost double that between the protein and the replaced water. The hydrogen bonding and van der Waals interactions are initially greater with the backbone and later with side chains of charged residues. Urea altered the water-water hydrogen bond network both by replacing water solvating hydrophobic residues and by shortening the first-shell intra-water hydrogen bonds by 0.2 Å. These interaction data suggest that urea uses both 'direct' and 'indirect' mechanisms to unfold lysozyme. Specific structural changes constitute the first steps in lysozyme unfolding by urea.