Particle Image Analysis Research Articles

Vesicle Picker: A tool for efficient identification of membrane protein complexes in vesicles

Electron cryomicroscopy (cryo-EM) has recently allowed determination of near-atomic resolution structures of membrane proteins and protein complexes embedded in lipid vesicles. However, particle selection from electron micrographs of these vesicles can be challenging due to the strong signal contributed from the lipid bilayer. This challenge often requires iterative and laborious particle selection workflows to generate a dataset of high-quality particle images for subsequent analysis. Here we present Vesicle Picker, an open-source program built on the Segment Anything model. Vesicle Picker enables automatic identification of vesicles in cryo-EM micrographs with high recall and precision. It then exhaustively selects all potential particle locations, either at the perimeter or uniformly over the surface of the projection of the vesicle. The program is designed to interface with cryoSPARC, which performs both upstream micrograph processing and downstream single particle image analysis. We demonstrate Vesicle Picker’s utility by determining a high-resolution map of the vacuolar-type ATPase from micrographs of native synaptic vesicles (SVs) and identifying an additional protein or protein complex in the SV membrane.

Molecular model of a bacterial flagellar motor in situ reveals a "parts-list" of protein adaptations to increase torque.

One hurdle to understanding how molecular machines work, and how they evolve, is our inability to see their structures in situ . Here we describe a minicell system that enables in situ cryogenic electron microscopy imaging and single particle analysis to investigate the structure of an iconic molecular machine, the bacterial flagellar motor, which spins a helical propeller for propulsion. We determine the structure of the high-torque Campylobacter jejuni motor in situ, including the subnanometre-resolution structure of the periplasmic scaffold, an adaptation essential to high torque. Our structure enables identification of new proteins, and interpretation with molecular models highlights origins of new components, reveals modifications of the conserved motor core, and explain how these structures both template a wider ring of motor proteins, and buttress the motor during swimming reversals. We also acquire insights into universal principles of flagellar torque generation. This approach is broadly applicable to other membrane-residing bacterial molecular machines complexes.

Particle formation in response to different protein formulations and containers: Insights from machine learning analysis of particle images

Subvisible particle count is a biotherapeutics stability indicator widely used by pharmaceutical industries. A variety of stresses that biotherapeutics are exposed to during development can impact particle morphology. By classifying particle morphological differences, stresses that have been applied to monoclonal antibodies (mAbs) can be identified. This study aims to evaluate common biotherapeutic drug storage and shipment conditions that are known to impact protein aggregation. Two different studies were conducted to capture particle images using micro-flow imaging and to classify particles using a convolutional neural network. The first study evaluated particles produced in response to agitation, heat, and freeze-thaw stresses in one mAb formulated in five different formulations. The second study evaluated particles from two common drug containers, a high-density polyethylene bottle and a glass vial, in six mAbs exposed solely to agitation stress. An extension of this study was also conducted to evaluate the impact of sequential stress exposure compared to exposure to one stress alone, on particle morphology. Overall, the convolutional neural network was able to classify particles belonging to a particular formulation or container. These studies indicate that storage and shipping stresses can impact particle morphology according to formulation composition and mAb.

Collapse Pattern in Gypseous Soil using Particle Image Velocimetry

Abstract Gypseous soil is prevalent in arid and semi-arid areas, is from collapsible soil, which contains the mineral gypsum, and has variable properties, including moisture-induced volume changes and solubility. Construction on these soils necessitates meticulous assessment and unique designs due to the possibility of foundation damage from soil collapse. The stability and durability of structures situated on gypseous soils necessitate close collaboration with specialists and careful, methodical preparation. It had not been done to find the pattern of failure in the micromechanical behavior of gypseous sandy soil through particle image velocity (PIV) analysis. This adopted recently in geotechnical engineering to track the motion of soil grains and using tracer particles by applying digital particle image analysis. It has also been used to study the displacement distribution in some cases of granular materials. Therefore, the goal of this study is to find out how gypseous sand medium moves when in contact with a rigid strip foundation that is under static stress and plane strain conditions. The experimental model would focus on two common types of wetting, namely water table rise and dry conditions. The PIV showed that the collapse pattern under the footing is of the type of punching shear failure. The predominant mechanism of soil deformation was the vertical compression of the gypseous granular soil. The results showed that understanding gypseous sandy grain displacement and failure patterns at the local scale is crucial for enhancing the design of foundations under static stress conditions.

PIV experimental study on water-sediment two-phase flow in NACA0012 cascade

Abstract Sediment wear is one of the important reasons for the damage of hydraulic machinery components, and the motion characteristics of sediment are the key factors to determine the degree of wear damage. In order to reveal the mechanism of sediment wear of turbine guide vane, motion characteristics of sediment near the wall of NACA0012 cascade were studied by PIV test method. In this research, fluorescent and sediment particles were used as tracer particles respectively, and the particle images of solid and liquid phases were recorded by double-camera synchronous imaging method. Finally, the synchronous velocity field of solid and liquid phases was determined by correlation analysis of particle images. The experimental results show that the PIV test method can accurately determine the motion characteristics of the solid phase. There is velocity slip between the solid phase and the liquid phase, and the velocity deviation is related to the factors of particle characteristics and Reynolds number.

Wear particle image analysis: feature extraction, selection and classification by deep and machine learning

PurposeThis study aims to explore the integration of machine learning (ML) in tribology to optimize lubrication interval decisions, aiming to enhance equipment lifespan and operational efficiency through wear image analysis.Design/methodology/approachUsing a data set of scanning electron microscopy images from an internal combustion engine, the authors used AlexNet as the feature extraction algorithm and the J48 decision tree algorithm for feature selection and compared 15 ML classifiers from the lazy-, Bayes and tree-based families.FindingsFrom the analyzed ML classifiers, instance-based k-nearest neighbor emerged as the optimal algorithm with a 95% classification accuracy against testing data. This surpassed individually trained convolutional neural networks’ (CNNs) and closely approached ensemble deep learning (DL) techniques’ accuracy.Originality/valueThe proposed approach simplifies the process, enhances efficiency and improves interpretability compared to more complex CNNs and ensemble DL techniques.

Automatic Particle Recognition Based on Digital lmage Processing

The purpose of the research is to develop and compare various methods and algorithms for effective particle analysis based on their visual characteristics. Тhe purpose of this article is to develop and compare various methods and algorithms for effective particle analysis based on their visual characteristics. Тhe paper considers two fundamentally different approaches: the analysis of grayscale gradients and the machine learning method.Methods.Тhe research methodology includes the analysis of particle images obtained by precipitation from colloidal solutions after laser ablation and images of powder particles for selective laser melting. Тhe materials were obtained using a Quanta 200 3D electron microscope (FЕ/). For the analysis, threshold brightness binarization, contour recognition methods by the Kenny operator and the Hough algorithm are used to combine boundary points into connected contours. For comparison, the U-Net neural network solution was used, and a dataset generator was created to train the neural network. Hand-cut images of aluminum alloy powder particles and micro and nanoparticles of various metals are used as data for generation.Results.Тhe results of the study show that the Hough method provides recognition of the number of particles at the level of 80%, and the machine learning method achieves 95% accuracy in recognizing the shape of particles. Both methods can be used to analyze microand nanoparticles, including irregularly shaped particles.Conclusion.Тhe findings of the work confirm that neural networks are the optimal solution for automatic particle recognition in digital images. However, in order to create a dataset of sufficient volume, it is necessary to develop a generator of labeled images, which requires a detailed study of the subject area.

Simcryocluster: a semantic similarity clustering method of cryo-EM images by adopting contrastive learning

BackgroundCryo-electron microscopy (Cryo-EM) plays an increasingly important role in the determination of the three-dimensional (3D) structure of macromolecules. In order to achieve 3D reconstruction results close to atomic resolution, 2D single-particle image classification is not only conducive to single-particle selection, but also a key step that affects 3D reconstruction. The main task is to cluster and align 2D single-grain images into non-heterogeneous groups to obtain sharper single-grain images by averaging calculations. The main difficulties are that the cryo-EM single-particle image has a low signal-to-noise ratio (SNR), cannot manually label the data, and the projection direction is random and the distribution is unknown. Therefore, in the low SNR scenario, how to obtain the characteristic information of the effective particles, improve the clustering accuracy, and thus improve the reconstruction accuracy, is a key problem in the 2D image analysis of single particles of cryo-EM.ResultsAiming at the above problems, we propose a learnable deep clustering method and a fast alignment weighted averaging method based on frequency domain space to effectively improve the class averaging results and improve the reconstruction accuracy. In particular, it is very prominent in the feature extraction and dimensionality reduction module. Compared with the classification method based on Bayesian and great likelihood, a large amount of single particle data is required to estimate the relative angle orientation of macromolecular single particles in the 3D structure, and we propose that the clustering method shows good results.ConclusionsSimcryoCluster can use the contrastive learning method to perform well in the unlabeled high-noise cryo-EM single particle image classification task, making it an important tool for cryo-EM protein structure determination

Digital Design of Crystallization Process: Why particle size and shape matters!

Populationbalance modelling is a widely used tool to understand the mechanism and kineticparameters of crystal growth at the bench, to facilitate scale-up to Production.In this work, we present recommendations for particle size and shape analyticalmethods to improve the predictive capability of 1D and 2D population balancemodels for crystallisation of contemporary active ingredients. For 1D models, wereport that the optical model for laser diffraction, a common technique formeasuring particle size distribution, can significantly impact the growth parametersresolved from the model. For 2D models, particle image analysis of powders andsingle crystals give valuable length and width data to predict how the aspectratio of particles evolves. Future work will be directed towards applying thesetools when scaling a crystallization process.

Experimental investigation on a novel external-mixing prefilming atomizer for advanced aircraft engines

As a crucial component of aircraft engine combustors, the external-mixing prefilming atomizer exhibits significant advantages in enhancing combustion efficiency and reducing pollutant emissions. However, the intricate relationship between spray characteristics, droplet breakup mechanisms, flow fields, and geometric parameters of the external-mixing prefilming atomizer remains poorly understood. This paper presents a novel external-mixing prefilming atomizer, aiming to address this knowledge gap. Additionally, particle image velocimetry (PIV), high-speed Schlieren photography, and particle image analysis (PIA) measuring systems were employed to gain insights into the spray characteristics and droplet breakup mechanism of the external-mixing prefilming atomizer under various swirl flow fields. The obtained results reveal the formation of two distinct regions, namely a low-density area and a high-density area. Notably, the number of droplets in the low-density area, located within 0–2 cm from the central axis, is significantly lower compared to the high-density area situated further away from the central axis. The breakup of droplets in the low-density region is primarily attributed to the inner swirl airflow, while the outer swirl airflow induces droplet breakup in the high-density region. As the Reynolds number of the incoming flow increases, the air-liquid ratio (ALR) and Weber number also increase, thereby enhancing the fuel breakup effect. The Schlieren results reveal a two-stage spray development process. The initial stage is characterized by the dominant momentum of the dense spray, with the spray tip penetration (STP) exhibiting a relatively slow increase. The aerodynamic force of airflow plays a crucial role in governing the breakup and transport of droplets, leading to a linear variation of STP over time. As the swirl number increases, there is an enhanced momentum exchange between droplets and airflow, resulting in accelerated spray penetration and droplet breakup.

Rab32, a novel Rab small GTPase from orange-spotted grouper, Epinephelus coioides involved in SGIV infection

Rab32 is a member of the Rab GTPase family that is involved in membrane trafficking and immune response, which are crucial for controlling pathogen infection. However, the role of Rab32 in virus infection is not well understood. In this study, we focused on the regulation of Rab32 on virus infection and the host immunity in orange-spotted grouper, Epinephelus coioides. EcRab32 encoded a 213-amino acid polypeptide, which shared a high sequence identity with other Rab32 proteins from fishes to mammals. In healthy orange-spotted grouper, the mRNA of EcRab32 was expressed in all the detected tissues, with the more expression levels in the head kidney, liver and gill. Upon SGIV infection, the expression of EcRab32 was significantly up-regulated in vitro, indicating its potential role in viral infection. EcRab32 was observed to be distributed in the cytoplasm as punctate and vesicle-like structures. EcRab32 overexpression was found to notably inhibit SGIV infection, while the interruption of EcRab32 significantly promoted SGIV infection. In addition, using single particle imaging analysis, we found that EcRab32 overexpression prominently reduced the attachment and internalization of SGIV particles. Furthermore, the results demonstrated that EcRab32 played a positive role in regulating the interferon immune and inflammatory responses. Taken together, these findings indicated that EcRab32 influenced SGIV infection by regulating the host immune response, providing an overall understanding of the interplay between the Rab32 and innate immunity.

Characteristics and Controlling Factors of Particle Crushing in Volcaniclastic Sediments under Compaction: The Quaternary Pyroclastic Deposits of Hongtu Hill, Changbaishan Area, Northeastern China

Recently, many scholars have conducted experimental mechanical compaction studies on sandstones, carbonates, and mudstones to visually study the mechanical compaction process and reservoir evolution of sedimentary rocks. However, experimental mechanical compaction studies on the evolution of the compaction process of pyroclastic rocks have been ignored. Volcaniclastic rock reservoirs are widely distributed across the world and strongly influenced by the crushing of pyroclastic particles. In this study, we analyzed the characteristics and controlling factors of the crushing of pyroclastic particles during compaction diagenesis from a microscopic perspective through experimental mechanical compaction. These results can provide quantitative compaction background parameters for the quantitative study of pyroclastic rock reservoir evolution. We took pyroclastic samples from Hongtu Hill in the Changbaishan area as an example, and experimental mechanical compaction experiments were conducted. Furthermore, image surface porosity and particle analysis statistical methods were used, and the variations in the effective porosity and image surface porosity under different axial stresses were studied. The results showed that, after compaction, the effective porosity did not exhibit a decreasing trend with increasing axial stress but rather a normal distribution trend that initially increased and then decreased. In the compaction experiment, the pyroclastic particle crushing process was segmented with increasing axial stress, and there was an obvious compaction band in the initial stage of the compaction, called the particle rearrangement stage (10–30 MPa). Furthermore, there were relatively non-successive compaction localization areas in the later stage of compaction, called the particle crushing stage (50–70 MPa), which was represented by vitreous basalt particles surrounded by porphyritic basalt particles. During experimental mechanical compaction, the smaller the compactness, the smaller the solidity, and the larger the slenderness of the particles, the more likely the particles were to break during compaction. Particles containing intragranular pores and vitreous basalt particles were easily crushed.

UTILE-Gen: Automated Image Analysis in Nanoscience Using Synthetic Dataset Generator and Deep Learning.

This work presents the development and implementation of a deep learning-based workflow for autonomous image analysis in nanoscience. A versatile, agnostic, and configurable tool was developed to generate instance-segmented imaging datasets of nanoparticles. The synthetic generator tool employs domain randomization to expand the image/mask pairs dataset for training supervised deep learning models. The approach eliminates tedious manual annotation and allows training of high-performance models for microscopy image analysis based on convolutional neural networks. We demonstrate how the expanded training set can significantly improve the performance of the classification and instance segmentation models for a variety of nanoparticle shapes, ranging from spherical-, cubic-, to rod-shaped nanoparticles. Finally, the trained models were deployed in a cloud-based analytics platform for the autonomous particle analysis of microscopy images.

Characteristics of Droplet Behaviors during Spray Breakup Process

The variation of droplet parameters during the spray breakup process affects the droplet deposition behavior and accurate application. The aim of this study was to experimentally investigate droplet behaviors along the penetration direction with respect to spray propagation. Particle image analysis (PIA) was applied to obtain the characteristics of droplets at three representative stages (namely, initial, quasi-steady, and end stages) after the start of injection (ASOI). The effects of timing and location on the spray characteristics were thoroughly investigated. First, different morphological changes of spray (droplets, ligaments, and bags) during spray breakup were observed. The experimental results show that droplet size and velocity distinctly increase from upstream to downstream at the initial stage. However, at the quasi-steady and end stages, droplet velocities are similar, and the effects of location are not evident. This indicates that location has a significant effect on droplet behaviors at the initial stage. The mean minimum distance (MD) of droplets first increases considerably and then decreases from upstream to downstream, suggesting that the droplets disperse better at midstream. Moreover, the mean MD at the initial stage exceeds that at the quasi-steady and end stages, denoting that the droplets disperse better with time. Finally, the geometric parameter of droplets and the key stage selection are important for predicting the interaction between the droplets and surfaces.

The influence of particle morphology on the long-term strength of sandy soil under freeze-thaw cycles

Freeze-thaw weathering can fragment sand and gravel materials, altering their mechanical properties. This study aims to simulate the effects of freeze-thaw weathering by subjecting sand and gravel materials to cycling tests. The tests were conducted in the monsoonal freeze zone, and the long-term strength and morphological indices of the sub-materials were measured after 0, 5, 10, 50, and 100 freeze-thaw cycles (FTCs) using a spherical template indenter and a particle image analyzer. The morphological indicators that were selected for this study – equivalent diameter and fractal dimension showed a strong correlation with long-term strength. The coefficients of determination (R) were 0.685 and 0.597, respectively. However, the roundness, aspect ratio, and ellipticity of the materials showed a weak correlation with long-term strength, with coefficients of determination less than 0.5. The freeze-thaw cycling process altered the degree of fragmentation of the sand and gravel materials, producing unique particle shape properties. This subsequently influences the alterations in the frictional resistance of the frozen soil, ultimately impacting the changes in its long-term strength. Improving the morphological characteristics of sand and gravel materials before use could be an effective way to enhance their resistance to freeze-thaw damage in engineering applications. In summary, our study demonstrates the importance of considering the effects of freeze-thaw weathering on the mechanical properties of sand and gravel materials. By improving their morphological characteristics, engineers can better protect these materials from damage caused by freeze-thaw cycling.

CFD-PBM Investigation on Droplet Size Distribution in a Liquid–Liquid Stirred Tank: Effect of Impeller Type

The influence of impeller type on the liquid–liquid dispersion system was investigated by computational fluid dynamics coupled with the population balance model (PBM). Rushton turbine (RT), pitched-blade turbine (PBT), and parabolic blade turbine (PB) produce different flow fields and turbulence distribution. The droplet size distribution (DSD), breakage frequency, and coalescence rate were predicted with the Alopaeus breakage model and the CT coalescence model. The simulated velocity field and DSD were validated with particle image velocimetry and image analysis, respectively. The analysis of the interaction between eddy and droplet shows that the turbulent inertial regime was the dominant mechanism of droplet breakage in the impeller region for radial-type impellers (RT and PB), while the turbulent viscous regime for an axial impeller (PBT). RT and PB can produce smaller droplets than PBT, about 16.4 and 10.7%, respectively. The average turbulent dissipation rates of RT and PB are about 7 and 10 times that of PBT in the impeller region, which explains the better performance of these two radial impellers in terms of droplet breakup.

Purification and CryoEM Image Analysis of the Bacterial Flagellar Filament.

The bacterial flagellum is a large assembly of about 30 different proteins and is divided into three parts: the filament that acts as a screw propeller, the hook as a universal joint, and the basal body as a rotary motor. In the case of Salmonella, the filament length is 10-15μm, which is more than ten times longer than the size of the cell. The filament is composed of only one component protein, flagellin, and is made of 11 protofilaments. The filament can form 12 different supercoiled structures as polymorphic forms. Each protofilament can take either the L (left-handed) or R (right-handed) state, and the number ratio of the protofilaments in these two states determines the shape of the supercoil. Some point mutations in flagellin make the filament straight by making all the protofilaments in one of the two states. The straight filaments enable us to use their helical symmetries for structural analysis by electron cryomicroscopy (cryoEM) and single particle image analysis. Here, we describe the methods for the purification of the flagellar filament and cryoEM data collection and image analysis.

Determination of the parameters of suspended particles of arbitrary shape by optoelectronic method

Data on the parameters of suspended particles are required in the electronic, optical, chemical and mining industries, powder metallurgy, atmospheric physics, medicine, etc. We present the results of determining the parameters of suspended particles of arbitrary shape using a combined optoelectronic approach, including a photoelectric method and a digital method of image processing. The results of both methods were used to correct the particle parameters and operational control of the laser emitter radiation. The operation of an optoelectronic device used to determine the parameters is based on the analysis of particle images in four projections, as well as on the registration of radiation scattered by particles and collected by an elliptical mirror. Spectrometry of the disperse composition of aerosols was carried out using a photomultiplier tube. The images of suspended particles for digital processing were obtained using a CCD array. It is shown that the applied image processing algorithms make it possible to remove noise, correct a background, improve the boundaries, determine the binarization threshold, exclude small particles, select closed areas and boundaries, split the image into the parts and count the number of particles in each of them, determine the parameters and classify the particles. Moreover, it becomes possible to restore the volumetric shape of a suspended particle and display the main characteristics on a digital indicator with an accuracy of 1%, which is significantly better than that of known similar devices. The results obtained can be used to control air pollution and improve the accuracy of identification of suspended particles of arbitrary shape.

Characterisation of iron oxide-containing pearlescent pigments used as food colourants: nano-labelling required in the EU?

Pearlescent pigments are used as colourants to increase the attractiveness of food products, especially in the patisserie and confectionery sector. They can be seen as composite materials and consist of thin potassium aluminium silicate (E 555, mica) platelets as carrier material, coated with a thin metal oxide layer of TiO2 (E 171) and/or iron oxides (E 172). The European Food Safety Authority stated in 2020 that mica-based pearlescent pigments as a whole should be evaluated as new food additives. Obtaining dependable data for particle size and layer thickness of these pigments is crucial both for the demanded food additive evaluation itself and also for the nanomaterial labelling assessment of products containing these food colourants according to the ‘Food Information to Consumers’ regulation. Since it was found in a previous study on TiO2-containing pearlescent pigments (silver and golden coloured) that the coating consisted of nanoscaled constituent titanium oxide particles, in this follow-up study we investigated whether Fe2O3-containing pearlescent pigments exhibit a similar nanostructured morphology. For this purpose, five commercially-available food products containing these pigments were investigated. Static light scattering and flow particle image analysis were used as screening methods to determine the mica platelet size. Scanning electron microscopy combined with energy-dispersive X-ray spectroscopy was used for nanostructure analysis of the metal oxide coating. The carrier mica platelets were 34–96 µm in diameter and 300–800 nm thick. The coating thickness was found to be in the range of 75–105 nm, with the constituent round shaped iron oxide particles contained therein having a minimum Feret diameter of 37–64 nm.

Eco-friendly Synthesis, Spectral Characterization, Particle image and Size Analysis of Cerium Oxide Nanoparticles Mediated By Waste Chow Chow Vegetable Seed Kernel Aqueous Extract

Eco-friendly Synthesis, Spectral Characterization, Particle image and Size Analysis of Cerium Oxide Nanoparticles Mediated By Waste Chow Chow Vegetable Seed Kernel Aqueous Extract