Automated model building and protein identification in cryo-EM maps.

Interpreting electron cryo-microscopy (cryo-EM) maps with atomic models requires high levels of expertise and labour-intensive manual intervention in three-dimensional computer graphics programs1,2. Here we present ModelAngelo, a machine-learning approach for automated atomic model building in cryo-EM maps. By combining information from the cryo-EM map with information from protein sequence and structure in a single graph neural network, ModelAngelo builds atomic models for proteins that are of similar quality to those generated by human experts. For nucleotides, ModelAngelo builds backbones with similar accuracy to those built by humans. By using its predicted amino acid probabilities for each residue in hidden Markov model sequence searches, ModelAngelo outperforms human experts in the identification of proteins with unknown sequences. ModelAngelo will therefore remove bottlenecks and increase objectivity in cryo-EM structure determination.

Atomic-resolution protein structure determination by cryo-EM.

Single-particle electron cryo-microscopy (cryo-EM) is a powerful method for solving the three-dimensional structures of biological macromolecules. The technological development of transmission electron microscopes, detectors and automated procedures in combination with user-friendly image processing software and ever-increasing computational power have made cryo-EM a successful and expanding technology over the past decade1. At resolutions better than 4 Å, atomic model building starts to become possible, but the direct visualization of true atomic positions in protein structure determination requires much higher (better than 1.5 Å) resolution, which so far has not been attained by cryo-EM. The direct visualization of atom positions is essential for understanding the mechanisms of protein-catalysed chemical reactions, and for studying how drugs bind to and interfere with the function of proteins2. Here we report a 1.25 Å-resolution structure of apoferritin obtained by cryo-EM with a newly developed electron microscope that provides, to our knowledge, unprecedented structural detail. Our apoferritin structure has almost twice the 3D information content of the current world record reconstruction (at 1.54 Å resolution3). We can visualize individual atoms in a protein, see density for hydrogen atoms and image single-atom chemical modifications. Beyond the nominal improvement in resolution, we also achieve a substantial improvement in the quality of the cryo-EM density map, which is highly relevant for using cryo-EM in structure-based drug design.

Single-particle cryo-EM at atomic resolution.

The three-dimensional positions of atoms in protein molecules define their structure and their roles in biological processes. The more precisely atomic coordinates are determined, the more chemical information can be derived and the more mechanistic insights into protein function may be inferred. Electron cryo-microscopy (cryo-EM) single-particle analysis has yielded protein structures with increasing levels of detail in recent years1,2. However, it has proved difficult to obtain cryo-EM reconstructions with sufficient resolution to visualize individual atoms in proteins. Here we use a new electron source, energy filter and camera to obtain a 1.7 Å resolution cryo-EM reconstruction for a human membrane protein, the ß3 GABAA receptor homopentamer3. Such maps allow a detailed understanding of small-molecule coordination, visualization of solvent molecules and alternative conformations for multiple amino acids, and unambiguous building of ordered acidic side chains and glycans. Applied to mouse apoferritin, our strategy led to a 1.22 Å resolution reconstruction that offers a genuine atomic-resolution view of a protein molecule using single-particle cryo-EM. Moreover, the scattering potential from many hydrogen atoms can be visualized in difference maps, allowing a direct analysis of hydrogen-bonding networks. Our technological advances, combined with further approaches to accelerate data acquisition and improve sample quality, provide a route towards routine application of cryo-EM in high-throughput screening of small molecule modulators and structure-based drug discovery.

How good can cryo-EM become?

The suddenness with which single-particle cryo-electron microscopy (cryo-EM) has emerged as a method for determining high-resolution structures of biological macromolecules invites the questions, how much better can this technology get, and how fast is that likely to happen? Though we can rightly celebrate the maturation of cryo-EM as a high-resolution structure-determination tool, I believe there still are many developments to look forward to.

Denoising of high-resolution single-particle electron-microscopy density maps by their approximation using three-dimensional Gaussian functions.

Cryo-electron microscopy (cryo-EM) of frozen-hydrated preparations of isolated macromolecular complexes is the method of choice to obtain the structure of complexes that cannot be easily studied by other experimental methods due to their flexibility or large size. An increasing number of macromolecular structures are currently being obtained at subnanometer resolution but the interpretation of structural details in such EM-derived maps is often difficult because of noise at these high-frequency signal components that reduces their contrast. In this paper, we show that the method for EM density-map approximation using Gaussian functions can be used for denoising of single-particle EM maps of high (typically subnanometer) resolution. We show its denoising performance using simulated and experimental EM density maps of several complexes.

The Electron Microscopy eXchange (EMX) initiative.

Three-dimensional electron microscopy (3DEM) of ice-embedded samples allows the structural analysis of large biological macromolecules close to their native state. Different techniques have been developed during the last forty years to process cryo-electron microscopy (cryo-EM) data. Not surprisingly, success in analysis and interpretation is highly correlated with the continuous development of image processing packages. The field has matured to the point where further progress in data and methods sharing depends on an agreement between the packages on how to describe common image processing tasks. Such standardization will facilitate the use of software as well as seamless collaboration, allowing the sharing of rich information between different platforms. Our aim here is to describe the Electron Microscopy eXchange (EMX) initiative, launched at the 2012 Instruct Image Processing Center Developer Workshop, with the intention of developing a first set of standard conventions for the interchange of information for single-particle analysis (EMX version 1.0). These conventions cover the specification of the metadata for micrograph and particle images, including contrast transfer function (CTF) parameters and particle orientations. EMX v1.0 has already been implemented in the Bsoft, EMAN, Xmipp and Scipion image processing packages. It has been and will be used in the CTF and EMDataBank Validation Challenges respectively. It is also being used in EMPIAR, the Electron Microscopy Pilot Image Archive, which stores raw image data related to the 3DEM reconstructions in EMDB.

Practical Experience with Hole-Free Phase Plates for Cryo Electron Microscopy.

Phase plate (PP) imaging has proven to be valuable in transmission cryo electron microscopy of unstained, native-state biological specimens. Many PP types have been described, however until the recent implementation of the "hole-free" phase plate (HFPP), imaging has been challenging. We found the HFPP to be simple to construct and to set up in the transmission electron microscopy, but care in implementing automated data collection is needed. Performance may be variable, both initially and over time, thus it is important to monitor and evaluate image quality by observing the power spectrum. We found that while some HFPPs gave transfer to high resolution without CTF oscillation, most reached high resolution when operated with modest defocus.

Automatic magnification determination of electron cryomicroscopy images using apoferritin as a standard.

Interpretation of the structural information in cryomicroscopy images recorded on film or CCD camera requires a precise knowledge of the electron microscope parameters that affect image features such as magnification and defocus. Magnification must be determined in order to combine data from different images in a three-dimensional reconstruction and to accurately scale reconstructions for fitting with atomic resolution models. A method is described for estimating the absolute magnification of an electron micrograph of a frozen-hydrated specimen using horse spleen apoferritin as a standard. Apoferritin is a widely available protein complex of known structure that may be included with the specimen of interest and imaged under conditions identical to those used for imaging other biological specimens by cryomicroscopy. The sum of the structure factor intensities of images of randomly-oriented apoferritin particles shows three low resolution peaks to 25Å that arise from the hollow ball structure of apoferritin. Comparison of peak positions of the experimental intensities with structure factor intensities of an atomic model of apoferritin determined by X-ray crystallography provides a scale factor for estimating the absolute magnification of the micrograph. We compare the magnification estimate using apoferritin to that obtained with tobacco mosaic virus, another common magnification standard for cryomicroscopy. We verify the precision of the method by acquiring images with a systematic variation of magnification.

MeasureIce: accessible on-the-fly measurement of ice thickness in cryo-electron microscopy.

Ice thickness is arguably one of the most important factors limiting the resolution of protein structures determined by cryo-electron microscopy (cryo-EM). The amorphous atomic structure of the ice that stabilizes and protects biological samples in cryo-EM grids also imprints some additional noise in cryo-EM images. Ice that is too thick jeopardizes the success of particle picking and reconstruction of the biomolecule in the worst case and, at best, deteriorates eventual map resolution. Minimizing the thickness of the ice layer and thus the magnitude of its noise contribution is thus imperative in cryo-EM grid preparation. In this paper we introduce MeasureIce, a simple, easy to use ice thickness measurement tool for screening and selecting acquisition areas of cryo-EM grids. We show that it is possible to simulate thickness-image intensity look-up tables, also usable in SerialEM and Leginon, using elementary scattering physics and thereby adapt the tool to any microscope without time consuming experimental calibration. We benchmark our approach using two alternative techniques: the "ice channel" technique and tilt-series tomography. We also demonstrate the utility of ice thickness measurement for selecting holes in gold grids containing an Equine apoferritin sample, achieving a 1.88 Ångstrom resolution in subsequent refinement of the atomic map.

Low-Dose Data Collection and Radiation Damage in MicroED.

Microcrystal electron diffraction (MicroED) is a technique for structure determination that relies on the strong interaction of electrons with a minuscule, crystalline sample. While some of the electrons used to probe the crystal interact without altering the crystal, others deposit energy which changes the sample through a series of damage events. It follows that the sample cannot be observed without damaging it, and the frames obtained at the beginning of data collection reflect a crystal that differs from the one that yields the last frames of the dataset. Data acquisition at cryogenic temperatures has been found to reduce the rate of damage progression and is routinely used to increase the dose tolerance of the crystal, allowing more useful data to be obtained before the sample is destroyed. Low-dose data collection can further prolong the lifetime of the crystal, such that less damage is inflicted over the course of data acquisition. Ideally, lower doses increase the measurable volume of a single-crystal lattice by reducing the damage caused by probing electrons. However, the information that can be recovered from a diffraction image is directly related to the number of electrons used to probe the sample. The signal from a weakly exposed crystal runs the risk of being lost in the noise contributed by solvent, crystal disorder, and the electron detection process. This work focuses on obtaining the best possible data from a MicroED measurement, which requires considering several aspects such as sample, dose, and camera type.

Improving particle quality in cryo-EM analysis using a PEGylation method.

Cryo-electron microscopy (cryo-EM) is widely used for structural biology studies and has been developed extensively in recent years. However, its sample vitrification process is a major limitation because it causes severe particle aggregation and/or denaturation. This effect is thought to occur because particles tend to stick to the "deadly" air-water interface during vitrification. Here, we report a method for PEGylation of proteins that can efficiently protect particles against such problems during vitrification. This method alleviates the laborious process of fine-tuning the vitrification conditions, allowing for analysis of samples that would otherwise be discarded.

High-Resolution Cryo-EM Structure of the Cardiac Actomyosin Complex.

Heart contraction depends on a complicated array of interactions between sarcomeric proteins required to convert chemical energy into mechanical force. Cyclic interactions between actin and myosin molecules, controlled by troponin and tropomyosin, generate the sliding force between the actin-based thin and myosin-based thick filaments. Alterations in this sophisticated system due to missense mutations can lead to cardiovascular diseases. Numerous structural studies proposed pathological mechanisms of missense mutations at the myosin-myosin, actin-tropomyosin, and tropomyosin-troponin interfaces. However, despite the central role of actomyosin interactions a detailed structural description of the cardiac actomyosin interface remained unknown. Here, we report a cryo-EM structure of a cardiac actomyosin complex at 3.8 Å resolution. The structure reveals the molecular basis of cardiac diseases caused by missense mutations in myosin and actin proteins.

Experimental Phasing of MicroED Data Using Radiation Damage.

We previously demonstrated that microcrystal electron diffraction (MicroED) can be used to determine atomic-resolution structures from vanishingly small three-dimensional crystals. Here, we present an example of an experimentally phased structure using only MicroED data. The structure of a seven-residue peptide is solved starting from differences to the diffraction intensities induced by structural changes due to radiation damage. The same wedge of reciprocal space was recorded twice by continuous-rotation MicroED from a set of 11 individual crystals. The data from the first pass were merged to make a "low-dose dataset." The data from the second pass were similarly merged to form a "damaged dataset." Differences between these two datasets were used to identify a single heavy-atom site from a Patterson difference map, and initial phases were generated. Finally, the structure was completed by iterative cycles of modeling and refinement.

High-Throughput Cryo-EM Enabled by User-Free Preprocessing Routines.

Single-particle cryoelectron microscopy (cryo-EM) continues to grow into a mainstream structural biology technique. Recent developments in data collection strategies alongside new sample preparation devices herald a future where users will collect multiple datasets per microscope session. To make cryo-EM data processing more automatic and user-friendly, we have developed an automatic pipeline for cryo-EM data preprocessing and assessment using a combination of deep-learning and image-analysis tools. We have verified the performance of this pipeline on a number of datasets and extended its scope to include sample screening by the user-free assessment of the qualities of a series of datasets under different conditions. We propose that our workflow provides a decision-free solution for cryo-EM, making data preprocessing more generalized and robust in the high-throughput era as well as more convenient for users from a range of backgrounds.

Benchmarking tomographic acquisition schemes for high-resolution structural biology.

Cryo electron tomography with subsequent subtomogram averaging is a powerful technique to structurally analyze macromolecular complexes in their native context. Although close to atomic resolution in principle can be obtained, it is not clear how individual experimental parameters contribute to the attainable resolution. Here, we have used immature HIV-1 lattice as a benchmarking sample to optimize the attainable resolution for subtomogram averaging. We systematically tested various experimental parameters such as the order of projections, different angular increments and the use of the Volta phase plate. We find that although any of the prominently used acquisition schemes is sufficient to obtain subnanometer resolution, dose-symmetric acquisition provides considerably better outcome. We discuss our findings in order to provide guidance for data acquisition. Our data is publicly available and might be used to further develop processing routines.

Membrane Protein Cryo-EM: Cryo-Grid Optimization and Data Collection with Protein in Detergent.

Cryo-electron microscopy (cryo-EM) is a powerful tool for investigating the structure of macromolecules under near-native conditions. Especially in the context of membrane proteins, this technique has allowed researchers to obtain structural information at a previously unattainable level of detail. Specimen preparation remains the bottleneck of most cryo-EM research projects, with membrane proteins representing particularly challenging targets of investigation due to their universal requirement for detergents or other solubilizing agents. Here we describe preparation of negative staining and cryo-EM grids and downstream data collection of membrane proteins in detergent, by far the most common solubilization agent. This protocol outlines a quick and straightforward procedure for screening and determining the structure of a membrane protein of interest under biologically relevant conditions.