Acta Crystallographica Section D Structural Biology

Calculation of density maps from atomic models is essential for structural studies using crystallography and electron cryo-microscopy (cryoEM). These maps serve various purposes, including atomic model building, refinement, visualization, and validation. However, accurately comparing model-calculated maps to experimental data poses challenges, particularly because the resolution of cryoEM experimental maps varies across the map. Traditional crystallography methods generate finite-resolution maps with uniform resolution throughout the unit cell volume, while most modern software in cryoEM employ Gaussian-like functions to generate these maps, which does not adequately account for atomic model parameters and resolution. Recent work by Urzhumtsev & Lunin (2022, IUCr Journal, 9, 728-734) introduces a novel method for computing atomic model maps that incorporate local resolution and can be expressed as analytically differentiable functions of all atomic parameters. This approach enhances the accuracy of matching atomic models to experimental maps. In this paper, we detail the implementation of this method in CCTBX and Phenix. SynopsisNew tools implemented in CCTBX and Phenix allow the calculation of variable-resolution maps through a sum of atomic images expressed as analytic functions of all atomic parameters, along with their associated local resolution.

We report the discovery of a new class of local minima that has severely limited the accuracy of macromolecular models. Termed density misfit barrier traps, these minima explain much of the poor fit between macromolecular models and experimental data relative to that of smaller molecules: not just high R factors, but distorted chemical geometry. We postulated that proteins exist as an ensemble of conformations that each have good geometry, but refinement algorithms have been unable to converge to them due to a tangling phenomenon arising from these traps. To demonstrate, a synthetic ground truth data set was generated, consisting of a 2-member ensemble with excellent geometry. A series of starting models, each trapped in increasingly difficult local minima, were prepared, a unified validation score defined, and an open Challenge issued. This Challenge inspired algorithms for escaping such traps, and new programs have been released that are expected to substantially improve the accuracy of macromolecular ensemble models. SynopsisA synthetic 2-member conformational ensemble of a small protein and corresponding electron density data was generated to demonstrate how topological local minima hinder simultaneous agreement with density data and chemical geometry restraints in conventional structure refinement.

Although X-ray crystallography captures the ensemble of conformations present within the crystal lattice, models typically depict only the most dominant conformation, obscuring the existence of alternative states. Applying the electron density-guided AlphaFold3 approach to {beta}2-Microglobulin highlights how ensembles of alternate backbone conformations can be systematically modeled directly from crystallographic maps. This study also highlights how the detection of conformational ensembles is affected by the local quality of electron density and subtle variations in crystallization conditions and lattice packing. These results demonstrate that density-guided AlphaFold3 can uncover conformational heterogeneity missed by conventional refinement, offering a robust, systematic framework to capture the full structural landscape of proteins in crystals and enhancing the interpretive power of macromolecular crystallography. SynopsisElectron-density-guided AlphaFold3 reveals previously unmodeled conformational heterogeneity in {beta}2-Microglobulin and shows how crystal packing influences ensemble detection in X-ray crystallography.

Monte Carlo neutron ray-tracing simulations of time-of-flight (TOF)-Laue neutron macromolecular crystal diffraction (n-MX) using the McStas software package were done for the upcoming NMX Macromolecular Diffractometer at the European Spallation Source. Splitting neutron rays that arrive at the crystal lead to dramatic improvements in event formation with minimal computational overhead. The simulated event probability data was sampled using a new single-pass weighted reservoir sampling method, and processed like real n-MX data using DIALS. The effects of air and beamstop scatter on simulated data was investigated. SynopsisMonte Carlo simulations of neutron protein diffraction experiments provide useful data that models instrumental components that interact with neutrons, as well as the crystal diffraction itself. These data can be applied to instrument development, such as the commissioning of the NMX Macromolecular Diffractometer at ESS.

Capturing meta-stable conformations of enzymes and ligand complexes demands structural snapshots beyond static crystal structures. While time-resolved serial crystallography at room temperature, offers a time-resolution down to the femto-second domain it requires large amounts of micro crystals, specialized beamlines and considerable experience. Moreover, as the majority of enzymes displays turnover-times in the millisecond domain or slower, simpler methods can provide meaningful structural insight into enzyme catalysis. Vitrification of protein crystals can trap reaction intermediates by rapid cooling to {inverted exclamation} 100 K, and has traditionally been used to gain insight into long lived reaction intermediates such as product complexes. However, manual vitrification procedures are limited to long delay times of at least several seconds and heavily suffer from operator variability. A solution to this problem is provided by automatic crystal plunging devices, such as the Spitrobot, that plunge loop-mounted protein crystals into liquid nitrogen within millisecond time-scales. Versatile means of reaction initiation can be achieved either by micro dispensing a ligand droplet, or via optical excitation of light-sensitive proteins, or via the photoactivation of caged compounds. In addition to the conceptual simplicity, another benefit of cryo-trapping is that data can be collected at conventional synchrotron beamlines, exploiting their robust high-throughput capabilities. Thus, compared to room-temperature time-resolved crystallography, users not only benefit from uncoupling sample-preparation and data-collection, but also from a reduction in the required technical expertise and ready access to radiation sources. However, as cryo-trapping crystallography explores dynamic structural changes that become only visible by the comparison of several samples, experiments have to be carefully planned to carry out the necessary controls and to avoid mis- or over-interpretation of the results. Here we describe a detailed protocol for cryo-trapping time-resolved crystallography using automated crystal-plungers that enables researchers to map enzymatic reaction coordinate pathways within the millisecond domain.

Filament-forming proteins such as TasA (Bacillus subtilis) and camelysins CalY1, CalY2 (Bacillus cereus) pose a particular challenge for structural analysis due to their strong tendency to self-association and their polydispersity, which severely limits their ability to crystallize or to be a target for NMR-spectroscopy. To address this, it is necessary to modify the amino acid sequence to prevent filamentation. Engineering a series of N- and C-terminal truncated variants by removing flexible parts is often key to success. N-terminal extensions are also a powerful tool for obtaining crystals of fiber-forming proteins.

Affinity chromatography is a powerful and therefore popular method for the purification of proteins for structural studies. The success of the technique relies on the specificity of the interaction between the target protein and the affinity resin. Here, we present the identification of two protein contaminants isolated from HEK293 cell lysate following affinity purification of twin Strep-tagged or FLAG-tagged proteins. The contaminants were identified as human propionyl-coenzyme A carboxylase (hPCC) and protein arginine methyltransferase 5 in complex with methylosome protein 50 (PRMT5:MEP50) via a combination of cryo-EM data processing and proteomic analyses. This report serves to illustrate how these contaminants may appear in cryo-EM datasets and to highlight the paramount importance of affinity chromatography resin specificity for efficient protein purification.

Microtubules are cytoskeletal filaments typically characterized by a discontinuous helical lattice of /{beta}-tubulin heterodimers. Microtubules can also adopt variable lattice architectures both in vitro and in cellular contexts. Pseudo-helical averaging processing strategies have been developed to generate cryo-EM reconstructions of microtubules with and without decorating protein-binding partners, but these pipelines can be difficult to implement for the average user, especially for undecorated filaments. Here, we describe MiCSPARC, a cryo-EM processing pipeline developed around CryoSPARC (Punjani et al., 2017), which leverages automated particle picking and fast 3D refinement times in CryoSPARC to determine structures of both decorated and undecorated microtubules. We generated reconstructions of undecorated GDP microtubules, as well as kinesin-1 motor domain-decorated GMPCPP filaments at resolutions of up to 2.8 [A], demonstrating the robustness of the pipeline. Based on its convenient implementation and ability to routinely generate high-resolution, seam-corrected microtubule reconstructions, MiCSPARC should provide a valuable tool for understanding microtubule dynamics, microtubule-associated proteins, and microtubule-targeting agents.

We present new tools for the structure determination of amyloid filaments from electron cryo-microscopy (cryo-EM) images. We introduce a new algorithm for automated filament picking, based on their characteristic 4.75 [A] repeat signal; we implement the new auto-picker in a fully automated procedure for on-the-fly pre-processing of cryo-EM data sets of amyloid filaments; we present a graphical tool to select filament types based on bi-hierarchical clustering of filaments and 2D class assignments; and we introduce a denoising neural network for Blush regularisation that is re-trained on amyloid reconstructions. The implementation of these tools in release 5.1 of our open-software package RELION ensures broad applicability. We demonstrate their usefulness on two experimental data sets, including a previously described data set on recombinant human islet amyloid protein (hIAPP) with the S20G mutation for which we identify two new filament types.

Cryogenic electron tomography (CryoET) provides 3D views of vitrified cellular samples, and protein structures can be determined from the tomograms by averaging many copies of the same protein computationally. However, the resolution of these averaged structures, particularly for smaller proteins, is often constrained by the precision of tilt-series alignment. In this study, we introduce a gradient descent-based approach to refine alignment parameters, enhancing the contrast in tomograms of the sample regions. This refinement not only improves contrast but also yields higher-resolution protein structures derived from the same particle populations.

Ligand binding has been shown to induce significant alterations in the conformational landscape of proteins. Traditional crystallography approaches have provided valuable input about the end states in ligand-binding reactions. However, dynamical relationships between ligand binding and backbone rearrangement often remain obscured by crystallographic structures. In the present study, we use time-resolved serial synchrotron crystallography (TR-SSX) to directly visualize indole binding in the cavity of T4 lysozyme L99A in microcrystals under controlled environmental conditions. By integrating fixed target crystallography with LAMA-based ligand delivery, we have been able to track the progression of ligand binding and backbone rearrangement. By utilizing an occupancy refinement protocol, we have been able to quantify structural populations. Our studies reveal that ligand binding for this protein cavity follows a diffusion-limited process that progressively rearranges the F -helix of the protein towards a dominant conformational state. These findings establish an observable link between ligand diffusion, occupancy evolution and conformational adaptation within a crystalline environment. More broadly, our work shows how TR-SSX can quantify ligand and conformational populations during binding, providing a framework to interpret structural adaptation in real time.

We develop a machine-automated approach for extracting results from papers, which we assess via a comprehensive review of the entire eLife corpus. Our method facilitates a direct comparison of machine and peer review, and sheds light on key challenges that must be overcome in order to facilitate AI-assisted science. In particular, the results point the way towards a machine-readable framework for disseminating scientific information. We therefore argue that publication systems should optimize separately for the dissemination of data and results versus the conveying of novel ideas, and the former should be machine-readable.

Fourier-space projection operations are central to electron microscopy single-particle analysis and electron tomography algorithms. Machine learning methods require differentiable implementations for end-to-end model training, but PyTorchs built-in operations are too slow for practical use. This paper introduces torch-projectors: a high-performance library for differentiable Fourier-space projections in PyTorch. The library provides 2D and 3D forward and backward projection operators with linear and cubic interpolation, supporting gradient calculation for all inputs. Optimized for CPU, Apple Silicon (MPS), and CUDA devices, torch-projectors outperforms torch-fourier-slice by 1-2 orders of magnitude.

Conventional two-dimensional (2D) histology relies upon destructive sample preparation and stereological estimation, frequently leading to sampling bias and loss of critical spatial context required for understanding renal structure relationships. Here, we detail a novel pipeline for high-resolution 3D histology of ex vivo murine kidneys using X-ray micro-computed tomography (micro-CT) at the high flux of a synchrotron light source, the architecture of the nephron and associated microvasculature necessitates three-dimensional (3D) analysis to accurately characterize its complexity. Soft-tissue contrast was optimized through an established phosphotungstic acid (PTA) staining protocol, enabling robust mapping of macro and microstructures via absorption contrast. Multi-scale imaging was performed, providing whole-organ context at resolutions around 3 m and achieving sub-micron detail (down to 400 nm) in targeted regions of interest (ROI) of the renal cortex. Utilizing machine learning segmentation pipelines optimized for large volumetric datasets, we extracted crucial 3D quantitative morphometric data. The results presented herein demonstrate accuracy and morphological insight achievable through synchrotron-based 3D imaging, establishing a robust method for quantitative preclinical research.

Plant-specific membrane receptor kinases with structurally diverse extracellular domains regulate key processes in plant growth, development, immunity and symbiosis. Structural studies of these glycoproteins are often hampered by the limited quantities in which they can be obtained. Here, we describe the LRR crystallization screen, which has enabled the successful crystallization and structure determination of multiple receptor kinase ectodomains, including ligand-and co-receptor-bound complexes. As an example, we report the 1.5 [A] resolution crystal structure of the leucine-rich repeat (LRR) domain of STRUBBELIG-RECEPTOR FAMILY 6 (SRF6) from Arabidopsis thaliana. The SRF6 ectodomain contains seven LRRs and a disulfide-bond-stabilised N-terminal capping domain but lacks the canonical C-terminal cap and the N-glycosylation pattern typically observed in other family members. Previously reported protein-protein interactions between the SRF6 and SRF7 ectodomains and the receptor kinases BRI1, BRL1, BRL3, SERK3 and BIR1-3 could not be confirmed by quantitative isothermal titration calorimetry and grating-coupled interferometry assays, suggesting that these structurally conserved LRR receptor kinases may have signalling functions outside the brassinosteroid pathway. SynopsisA crystallisation screen that has enabled the structural analysis of various extracellular domains of plant membrane receptor kinases is described together.

AbstractThe development of machine learning models for protein-ligand interactions is fundamentally constrained by the quality and diversity of available structural data. Existing databases of protein-ligand complexes present researchers with an unsatisfying trade-off: carefully curated collections such as PDBBind and HiQBind offer high structural reliability but cover only a narrow slice of the Protein Data Bank (PDB), while large-scale resources like PLInder provide broad coverage at the expense of rigorous quality control. Here, we introduce CROWN (Curated Repository Of Well-resolved Non-covalent interactions), a machine learning-ready dataset that reconciles this tension by applying a comprehensive, fully automated preprocessing pipeline to the PLInder database. Starting from 649,915 protein-ligand interaction systems, CROWN applies a series of interleaved quality filters and processing stages addressing crystallographic resolution, ligand identity, pocket completeness, structural repair, interaction quality, and protonation at physiological pH. A distinguishing feature of the pipeline is a final constrained energy minimisation step using custom flat-bottomed restraints, which balances crystallographic evidence with relaxation of intramolecular strain. This step -- absent from existing protein-ligand datasets -- produces structurally uniform complexes by reconciling the heterogeneous refinement practices of different crystallographers and structure determination protocols, without distorting the experimentally observed binding geometry. The resulting dataset of 153,005 complexes represents a roughly four-fold increase in protein and species diversity over PDBBind and HiQBind, while maintaining rigorous structural standards. Importantly, CROWN adopts a geometry-centric design philosophy that treats the 3D arrangement of atoms at the binding interface as a self-consistent source of information, rather than relying on externally measured binding affinities that cover only a fraction of known structures and introduce well-documented biases. We anticipate that CROWN will serve as a broadly useful resource for training generative models of protein-ligand binding poses, developing scoring functions, and benchmarking interaction prediction methods.

Multi-domain proteins (MDPs) adopt diverse conformations arising from cooperative inter-domain motions, and such dynamics are intimately coupled to their biological functions. Quantitative characterization of these motions is crucial for elucidating their functional mechanisms. Although small-angle X-ray scattering (SAXS) provides information on overall domain arrangement, the limited experimental constraints hinder reliable discrimination of conformational ensembles derived from molecular dynamics (MD) simulations. To address this limitation, complementary experimental constraints that enable to observe domain-selective structural information are required. Inverse contrast-matching small-angle neutron scattering (iCM-SANS), combined with segmental deuteration, enables selective visualization of individual domains and thus provides such complementary information. However, practical strategies for preparing segmentally deuterated MDPs with arbitrary domain labelling have yet to be established. Here, we develop an experimental protocol that integrates controlled protein deuteration with high-efficiency multi-step protein ligation to generate a segmentally deuterated MDP in high yield. The combined use of SAXS and iCM-SANS yields complementary structural constraints that enhance discrimination of MD-derived conformational ensembles. This protocol expands the applicability of segment-selective visualization and also provides an opportunity for high-precision analysis of dynamics in complex MDPs. SynopsisSegmental deuteration enabled by high-efficiency multi-step protein ligation, combined with inverse contrast-matching SANS and SAXS, provides structural constraints that improve discrimination of molecular dynamics ensembles of multi-domain proteins. IMPORTANTthis document contains embedded data - to preserve data integrity, please ensure where possible that the IUCr Word tools (available from http://journals.iucr.org/services/docxtemplate/) are installed when editing this document.

Glutamate racemase (MurI) catalyzes the stereochemical interconversion of L-glutamate to D-glutamate, a key element of bacterial peptidoglycan biosynthesis. In this study, we present the crystal structure of Helicobacter pylori glutamate racemase at 1.43 [A] and in monoclinic symmetry, as previously reported models, but different unit-cell parameters. The present model contains a single dimer and retains the previously described head-to-head dimer arrangement. The differences between the models arise from variations in unit-cell parameters, which lead to altered crystal packing interactions rather than changes in the quaternary assembly. The monomeric fold and active-site architecture remain conserved and are consistent with the catalytic features described for bacterial glutamate racemases. This structure provides an updated, high-resolution structural model for H. pylori glutamate racemase and highlights the variability in crystal packing within the same space group.

Multidomain proteins connected by flexible linkers populate conformational ensembles that are challenging to characterize using conventional structural biology methods. In domain-linker-domain (DLD) proteins, linker-mediated inter-domain relative positions and orientations are functionally relevant, yet their dynamical behavior in solution normally remain poorly described. Small-angle X-ray scattering (SAXS) provides ensemble-averaged structural information for such systems; however, coupling with computational modeling is required to accurately describe the dynamic behavior of this family of proteins in solution. Here, we present a systematic evaluation of five ensemble-generation strategies applied to a set of eighteen proteins sharing the same two globular domains, connected by naturally occurring linkers of varying length and composition. Modeling methods based on different underlying principles are compared by assessing their agreement to experimental SAXS data, showing a large disparity and systematic structural biases among them. Furthermore, for each approach, we examine the effect of refinement against SAXS restraints and assess its capacity to describe the experimental data, as well as the induced biases in global dimensions and inter-domain distance distributions. This analysis underlines the importance of the initial conformational pool for deriving experimentally compatible ensembles. Overall, this work provides a high-quality benchmark for SAXS-driven ensemble modeling of flexible, multidomain proteins and establishes a framework for the critical interpretation of solution scattering data in systems with pronounced conformational heterogeneity.

We report on recent methodological advances at the Life Science X-ray Scattering (LiX) beamline of the National Synchrotron Light Source II (NSLS-II) to support small-angle X-ray scattering experiments on skeletal and cardiac muscle tissues. These experiments have been routinely performed at the BioCAT beamline of the Advanced Photon Source (APS) over the past two decades to measure sarcomeric protein organization within healthy and diseased muscle tissues and provide direct molecular evidence for their functional roles and dynamics. Many recent advances in our understanding of sarcomeric proteins relied on diffraction data and include, as examples, MyBP-C, crossbridge SRX/DRX states, and titin. With LiX now available for muscle experimentation, more muscle users can be supported which will speed up research of sarcomeric proteins, muscle biomechanics, and skeletal and cardiac myopathies. LiX explicitly focuses on high-throughput muscle diffraction with rapid sample turnover and semi-automated data processing. These operations have been tested and validated on skeletal and cardiac tissues sourced from both humans and multiple animal models including pig, rat, mouse, and zebrafish.