Laser-induced metastable mixed phase of AuNi nanoparticles: a coherent X-ray diffraction imaging study.

The laser annealing process for AuNi nanoparticles has been visualized using coherent X-ray diffraction imaging (CXDI). AuNi bimetallic alloy nanoparticles, originally phase separated due to the miscibility gap, transform to metastable mixed alloy particles with rounded surface as they are irradiated by laser pulses. A three-dimensional CXDI shows that the internal part of the AuNi particles is in the mixed phase with preferred compositions at ∼29 at% of Au and ∼90 at% of Au.

Pixel detectors for diffraction-limited storage rings.

Dramatic advances in synchrotron radiation sources produce ever-brighter beams of X-rays, but those advances can only be used if there is a corresponding improvement in X-ray detectors. With the advent of storage ring sources capable of being diffraction-limited (down to a certain wavelength), advances in detector speed, dynamic range and functionality is required. While many of these improvements in detector capabilities are being pursued now, the orders-of-magnitude increases in brightness of diffraction-limited storage ring sources will require challenging non-incremental advances in detectors. This article summarizes the current state of the art, developments underway worldwide, and challenges that diffraction-limited storage ring sources present for detectors.

SpeckleNN: a unified embedding for real-time speckle pattern classification in X-ray single-particle imaging with limited labeled examples.

With X-ray free-electron lasers (XFELs), it is possible to determine the three-dimensional structure of noncrystalline nanoscale particles using X-ray single-particle imaging (SPI) techniques at room temperature. Classifying SPI scattering patterns, or `speckles', to extract single-hits that are needed for real-time vetoing and three-dimensional reconstruction poses a challenge for high-data-rate facilities like the European XFEL and LCLS-II-HE. Here, we introduce SpeckleNN, a unified embedding model for real-time speckle pattern classification with limited labeled examples that can scale linearly with dataset size. Trained with twin neural networks, SpeckleNN maps speckle patterns to a unified embedding vector space, where similarity is measured by Euclidean distance. We highlight its few-shot classification capability on new never-seen samples and its robust performance despite having only tens of labels per classification category even in the presence of substantial missing detector areas. Without the need for excessive manual labeling or even a full detector image, our classification method offers a great solution for real-time high-throughput SPI experiments.

Unsupervised learning approaches to characterizing heterogeneous samples using X-ray single-particle imaging.

One of the outstanding analytical problems in X-ray single-particle imaging (SPI) is the classification of structural heterogeneity, which is especially difficult given the low signal-to-noise ratios of individual patterns and the fact that even identical objects can yield patterns that vary greatly when orientation is taken into consideration. Proposed here are two methods which explicitly account for this orientation-induced variation and can robustly determine the structural landscape of a sample ensemble. The first, termed common-line principal component analysis (PCA), provides a rough classification which is essentially parameter free and can be run automatically on any SPI dataset. The second method, utilizing variation auto-encoders (VAEs), can generate 3D structures of the objects at any point in the structural landscape. Both these methods are implemented in combination with the noise-tolerant expand-maximize-compress (EMC) algorithm and its utility is demonstrated by applying it to an experimental dataset from gold nanoparticles with only a few thousand photons per pattern. Both discrete structural classes and continuous deformations are recovered. These developments diverge from previous approaches of extracting reproducible subsets of patterns from a dataset and open up the possibility of moving beyond the study of homogeneous sample sets to addressing open questions on topics such as nanocrystal growth and dynamics, as well as phase transitions which have not been externally triggered.

Erratum: An advanced workflow for single-particle imaging with the limited data at an X-ray free-electron laser. Corrigendum.

[This corrects the article DOI: 10.1107/S2052252520012798.].

A Metropolis Monte Carlo algorithm for merging single-particle diffraction intensities.

Single-particle imaging with X-ray free-electron lasers depends crucially on algorithms that merge large numbers of weak diffraction patterns despite missing measurements of parameters such as particle orientations. The expand-maximize-compress (EMC) algorithm is highly effective at merging single-particle diffraction patterns with missing orientation values, but most implementations exhaustively sample the space of missing parameters and may become computationally prohibitive as the number of degrees of freedom extends beyond orientation angles. This paper describes how the EMC algorithm can be modified to employ Metropolis Monte Carlo sampling rather than grid sampling, which may be favorable for reconstruction problems with more than three missing parameters. Using simulated data, this variant is compared with the standard EMC algorithm.

Noise reduction and mask removal neural network for X-ray single-particle imaging.

Free-electron lasers could enable X-ray imaging of single biological macromolecules and the study of protein dynamics, paving the way for a powerful new imaging tool in structural biology, but a low signal-to-noise ratio and missing regions in the detectors, colloquially termed 'masks', affect data collection and hamper real-time evaluation of experimental data. In this article, the challenges posed by noise and masks are tackled by introducing a neural network pipeline that aims to restore diffraction intensities. For training and testing of the model, a data set of diffraction patterns was simulated from 10 900 different proteins with molecular weights within the range of 10-100 kDa and collected at a photon energy of 8 keV. The method is compared with a simple low-pass filtering algorithm based on autocorrelation constraints. The results show an improvement in the mean-squared error of roughly two orders of magnitude in the presence of masks compared with the noisy data. The algorithm was also tested at increasing mask width, leading to the conclusion that demasking can achieve good results when the mask is smaller than half of the central speckle of the pattern. The results highlight the competitiveness of this model for data processing and the feasibility of restoring diffraction intensities from unknown structures in real time using deep learning methods. Finally, an example is shown of this preprocessing making orientation recovery more reliable, especially for data sets containing very few patterns, using the expansion-maximization-compression algorithm.

Experimental 3D coherent diffractive imaging from photon-sparse random projections.

The routine atomic resolution structure determination of single particles is expected to have profound implications for probing structure-function relationships in systems ranging from energy-storage materials to biological molecules. Extremely bright ultrashort-pulse X-ray sources - X-ray free-electron lasers (XFELs) - provide X-rays that can be used to probe ensembles of nearly identical nanoscale particles. When combined with coherent diffractive imaging, these objects can be imaged; however, as the resolution of the images approaches the atomic scale, the measured data are increasingly difficult to obtain and, during an X-ray pulse, the number of photons incident on the 2D detector is much smaller than the number of pixels. This latter concern, the signal 'sparsity', materially impedes the application of the method. An experimental analog using a conventional X-ray source is demonstrated and yields signal levels comparable with those expected from single biomolecules illuminated by focused XFEL pulses. The analog experiment provides an invaluable cross check on the fidelity of the reconstructed data that is not available during XFEL experiments. Using these experimental data, it is established that a sparsity of order 1.3 × 10-3 photons per pixel per frame can be overcome, lending vital insight to the solution of the atomic resolution XFEL single-particle imaging problem by experimentally demonstrating 3D coherent diffractive imaging from photon-sparse random projections.

Common structural features of toxic intermediates from α-synuclein and GroES fibrillogenesis detected using cryogenic coherent X-ray diffraction imaging.

The aggregation and deposition of α-synuclein (αSyn) in neuronal cells is correlated to pathogenesis of Parkinson's disease. Although the mechanism of αSyn aggregation and fibril formation has been studied extensively, the structural hallmarks that are directly responsible for toxicity toward cells are still under debate. Here, we have compared the structural characteristics of the toxic intermediate molecular species of αSyn and similar toxic species of another protein, GroES, using coherent X-ray diffraction analysis. Using coherent X-ray free electron laser pulses of SACLA, we analysed αSyn and GroES fibril intermediate species and characterized various aggregate structures. Unlike previous studies where an annular oligomeric form of αSyn was identified, particle reconstruction from scattering traces suggested that the specific forms of the toxic particles were varied, with the sizes of the particles falling within a specific range. We did however discover a common structural feature in both αSyn and GroES samples; the edges of the detected particles were nearly parallel and produced a characteristic diffraction pattern in the diffraction experiments. The presence of parallel-edged particles in toxic intermediates of αSyn and GroES fibrillogenesis pointed towards a plausible common molecular interface that leads to the formation of mature fibrils.

Analysis of XFEL serial diffraction data from individual crystalline fibrils.

Serial diffraction data collected at the Linac Coherent Light Source from crystalline amyloid fibrils delivered in a liquid jet show that the fibrils are well oriented in the jet. At low fibril concentrations, diffraction patterns are recorded from single fibrils; these patterns are weak and contain only a few reflections. Methods are developed for determining the orientation of patterns in reciprocal space and merging them in three dimensions. This allows the individual structure amplitudes to be calculated, thus overcoming the limitations of orientation and cylindrical averaging in conventional fibre diffraction analysis. The advantages of this technique should allow structural studies of fibrous systems in biology that are inaccessible using existing techniques.

Digital renovation of S2001 field X-ray vehicle / 医疗卫生装备

S2001 field X-ray vehicle is used to perform X-ray examination in the field hospital, while the disadvantages of the traditional film-screen radiography restrict its role of field examination in this digital era. By analyzing the detail of modern digital X-ray radiography, and comparing three different digital X-ray flat detectors, a solution is put forward to update S2001 field X-ray vehicle to a field digital X- ray radiography vehicle by using Canon CXDI digital X- ray detector.

Digital Renovation of Field X-ray Vehicle / 医疗卫生装备

Objective To upgrade the current X-ray vehicle.Methods Modern digital X-ray radiography was analyzed and 3 kinds of X-ray detectors were compared.Canon CXDI digital X-ray detector was used to update field X-ray vehicle into a field digital X-ray radiography vehicle.Results Real-time X-ray signal synchronization was realized.Digital images could be edited and labeled with detection information.Conclusion The working efficiency is enhanced and the support ability is improved.