Cell Imaging Data Research Articles

Topological properties accurately predict cell division events and organization of shoot apical meristem in Arabidopsis thaliana.

ABSTRACTCell division and the resulting changes to the cell organization affect the shape and functionality of all tissues. Thus, understanding the determinants of the tissue-wide changes imposed by cell division is a key question in developmental biology. Here, we use a network representation of live cell imaging data from shoot apical meristems (SAMs) in Arabidopsis thaliana to predict cell division events and their consequences at the tissue level. We show that a support vector machine classifier based on the SAM network properties is predictive of cell division events, with test accuracy of 76%, which matches that based on cell size alone. Furthermore, we demonstrate that the combination of topological and biological properties, including cell size, perimeter, distance and shared cell wall between cells, can further boost the prediction accuracy of resulting changes in topology triggered by cell division. Using our classifiers, we demonstrate the importance of microtubule-mediated cell-to-cell growth coordination in influencing tissue-level topology. Together, the results from our network-based analysis demonstrate a feedback mechanism between tissue topology and cell division in A. thaliana SAMs.

Mathematical model and genomics construction of developmental biology patterns using digital image technology.

Biological pattern formation ensures that tissues and organs develop in the correct place and orientation within the body. A great deal has been learned about cell and tissue staining techniques, and today’s microscopes can capture digital images. A light microscope is an essential tool in biology and medicine. Analyzing the generated images will involve the creation of unique analytical techniques. Digital images of the material before and after deformation can be compared to assess how much strain and displacement the material responds. Furthermore, this article proposes Development Biology Patterns using Digital Image Technology (DBP-DIT) to cell image data in 2D, 3D, and time sequences. Engineered materials with high stiffness may now be characterized via digital image correlation. The proposed method of analyzing the mechanical characteristics of skin under various situations, such as one direction of stress and temperatures in the hundreds of degrees Celsius, is achievable using digital image correlation. A DBP-DIT approach to biological tissue modeling is based on digital image correlation (DIC) measurements to forecast the displacement field under unknown loading scenarios without presupposing a particular constitutive model form or owning knowledge of the material microstructure. A data-driven approach to modeling biological materials can be more successful than classical constitutive modeling if adequate data coverage and advice from partial physics constraints are available. The proposed procedures include a wide range of biological objectives, experimental designs, and laboratory preferences. The experimental results show that the proposed DBP-DIT achieves a high accuracy ratio of 99,3%, a sensitivity ratio of 98.7%, a specificity ratio of 98.6%, a probability index of 97.8%, a balanced classification ratio of 97.5%, and a low error rate of 38.6%.

Segmentation, tracking and cell cycle analysis of live-cell imaging data with Cell-ACDC

BackgroundHigh-throughput live-cell imaging is a powerful tool to study dynamic cellular processes in single cells but creates a bottleneck at the stage of data analysis, due to the large amount of data generated and limitations of analytical pipelines. Recent progress on deep learning dramatically improved cell segmentation and tracking. Nevertheless, manual data validation and correction is typically still required and tools spanning the complete range of image analysis are still needed.ResultsWe present Cell-ACDC, an open-source user-friendly GUI-based framework written in Python, for segmentation, tracking and cell cycle annotations. We included state-of-the-art deep learning models for single-cell segmentation of mammalian and yeast cells alongside cell tracking methods and an intuitive, semi-automated workflow for cell cycle annotation of single cells. Using Cell-ACDC, we found that mTOR activity in hematopoietic stem cells is largely independent of cell volume. By contrast, smaller cells exhibit higher p38 activity, consistent with a role of p38 in regulation of cell size. Additionally, we show that, in S. cerevisiae, histone Htb1 concentrations decrease with replicative age.ConclusionsCell-ACDC provides a framework for the application of state-of-the-art deep learning models to the analysis of live cell imaging data without programming knowledge. Furthermore, it allows for visualization and correction of segmentation and tracking errors as well as annotation of cell cycle stages. We embedded several smart algorithms that make the correction and annotation process fast and intuitive. Finally, the open-source and modularized nature of Cell-ACDC will enable simple and fast integration of new deep learning-based and traditional methods for cell segmentation, tracking, and downstream image analysis.Source code: https://github.com/SchmollerLab/Cell_ACDC

Towards 'end-to-end' analysis and understanding of biological timecourse data.

Petabytes of increasingly complex and multidimensional live cell and tissue imaging data are generated every year. These videos hold large promise for understanding biology at a deep and fundamental level, as they capture single-cell and multicellular events occurring over time and space. However, the current modalities for analysis and mining of these data are scattered and user-specific, preventing more unified analyses from being performed over different datasets and obscuring possible scientific insights. Here, we propose a unified pipeline for storage, segmentation, analysis, and statistical parametrization of live cell imaging datasets.

Abstract 3106: Identifying drivers in the converging syntenic aneuploidies of spontaneous canine and pediatric high-grade glioma using imaging-based an arrayed CRISPR-Cas9 phenotypic screen

Abstract Gliomas occur in companion dogs at rates comparable to humans, with short-snouted breeds such as boxers being more susceptible than others. The natural progression of cancer in the immuno-competent host allows companion dogs diagnosed with sporadic glioma as an optimal model for preclinical testing of therapeutic approaches with human relevance, including immunotherapies. We have recently performed comprehensive genomic and epigenetic characterization of glioma in dogs to their human counterparts and found strong convergent evolution - shared somatic mutations and aneuploidies - among syntenic regions, including those of known pediatric glioma drivers, e.g., PDGFRA, MYC, PIK3CA. Here, using arrayed CRISPR-Cas9 imaging based phenotypic screen, we will probe potential oncogenic drivers and tumor suppressor genes within syntenic aneuploidies and thus outline functional versus non-functional heterogeneity of cancer aneuploidy. Specifically, we are conducting arrayed knockout screen (one gene per well) of 400+ genes within syntenic aneuploidies across primary cultured cells of canine glioma (n=2) and pediatric high-grade glioma cell lines (n=2). We will first capture images by high-speed confocal imaging system at three time points post-transduction of single guide RNAs (2 per gene) targeting each of 400+ genes in their separate wells. Then, using high-throughput image analysis and semi-supervised machine learning methods, we will measure well-based phenotypic features (viability, growth, and morphology) from these images. Genes will be ranked per cross-validated predicted probability in yielding either proliferating or slow-growing cell type based on learned phenotypic features using image data of knockout cells from and across wells. The top ranked genes will then be linked to oncogenes and tumor suppressors based on pathway and ontology analysis as well as further functional in vitro and in vivo (PDX) validation. We expect to find convergence of the most impactful molecular abnormalities (based on their knockout phenotypes) on candidate signaling pathways for the development of new drugs and repurposing of existing drugs for children and dogs with high-grade glioma. Citation Format: Samirkumar B. Amin, Amit Gujar, Eunhee Yi, Wonyeong Kang, Megan Costa, Greg Sjogren, Paul Gabriel, Leigh Maher, Peter Dickinson, Rebecca Packer, Elise Courtois, Paul Robson, Charles Lee, Roel Verhaak. Identifying drivers in the converging syntenic aneuploidies of spontaneous canine and pediatric high-grade glioma using imaging-based an arrayed CRISPR-Cas9 phenotypic screen [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 3106.

SPF: A spatial and functional data analytic approach to cell imaging data.

The tumor microenvironment (TME), which characterizes the tumor and its surroundings, plays a critical role in understanding cancer development and progression. Recent advances in imaging techniques enable researchers to study spatial structure of the TME at a single-cell level. Investigating spatial patterns and interactions of cell subtypes within the TME provides useful insights into how cells with different biological purposes behave, which may consequentially impact a subject's clinical outcomes. We utilize a class of well-known spatial summary statistics, the K-function and its variants, to explore inter-cell dependence as a function of distances between cells. Using techniques from functional data analysis, we introduce an approach to model the association between these summary spatial functions and subject-level outcomes, while controlling for other clinical scalar predictors such as age and disease stage. In particular, we leverage the additive functional Cox regression model (AFCM) to study the nonlinear impact of spatial interaction between tumor and stromal cells on overall survival in patients with non-small cell lung cancer, using multiplex immunohistochemistry (mIHC) data. The applicability of our approach is further validated using a publicly available multiplexed ion beam imaging (MIBI) triple-negative breast cancer dataset.

Temporal and Locational Values of Images Affecting the Deep Learning of Cancer Stem Cell Morphology.

Deep learning is being increasingly applied for obtaining digital microscopy image data of cells. Well-defined annotated cell images have contributed to the development of the technology. Cell morphology is an inherent characteristic of each cell type. Moreover, the morphology of a cell changes during its lifetime because of cellular activity. Artificial intelligence (AI) capable of recognizing a mouse-induced pluripotent stem (miPS) cell cultured in a medium containing Lewis lung cancer (LLC) cell culture-conditioned medium (cm), miPS-LLCcm cell, which is a cancer stem cell (CSC) derived from miPS cell, would be suitable for basic and applied science. This study aims to clarify the limitation of AI models constructed using different datasets and the versatility improvement of AI models. The trained AI was used to segment CSC in phase-contrast images using conditional generative adversarial networks (CGAN). The dataset included blank cell images that were used for training the AI but they did not affect the quality of predicting CSC in phase contrast images compared with the dataset without the blank cell images. AI models trained using images of 1-day culture could predict CSC in images of 2-day culture; however, the quality of the CSC prediction was reduced. Convolutional neural network (CNN) classification indicated that miPS-LLCcm cell image classification was done based on cultivation day. By using a dataset that included images of each cell culture day, the prediction of CSC remains to be improved. This is useful because cells do not change the characteristics of stem cells owing to stem cell marker expression, even if the cell morphology changes during culture.

Deep learning-based subcellular phenotyping of protrusion dynamics reveals fine differential drug responses at subcellular and single-cell levels

Intracellular processes such as cytoskeletal organization and organelle dynamics exhibit massive subcellular heterogeneity. Although recent advances in fluorescence microscopy allow researchers to acquire an unprecedented amount of live cell image data at high spatiotemporal resolutions, the traditional ensemble-averaging of uncharacterized subcellular heterogeneity could mask important activities. Moreover, the curse of dimensionality of these complex dynamic datasets prevents access to critical mechanistic details of subcellular processes.

Different cell imaging methods did not significantly improve immune cell image classification performance.

Developments in high-throughput microscopy have made it possible to collect huge amounts of cell image data that are difficult to analyse manually. Machine learning (e.g., deep learning) is often employed to automate the extraction of information from these data, such as cell counting, cell type classification and image segmentation. However, the effects of different imaging methods on the accuracy of image processing have not been examined systematically. We studied the effects of different imaging methods on the performance of machine learning-based cell type classifiers. We observed lymphoid-primed multipotential progenitor (LMPP) and pro-B cells using three imaging methods: differential interference contrast (DIC), phase contrast (Ph) and bright-field (BF). We examined the classification performance of convolutional neural networks (CNNs) with each of them and their combinations. CNNs achieved an area under the receiver operating characteristic (ROC) curve (AUC) of ~0.9, which was significantly better than when the classifier used only cell size or cell contour shape as input. However, no significant differences were found between imaging methods and focal positions.

A deep generative model of 3D single-cell organization

We introduce a framework for end-to-end integrative modeling of 3D single-cell multi-channel fluorescent image data of diverse subcellular structures. We employ stacked conditional β-variational autoencoders to first learn a latent representation of cell morphology, and then learn a latent representation of subcellular structure localization which is conditioned on the learned cell morphology. Our model is flexible and can be trained on images of arbitrary subcellular structures and at varying degrees of sparsity and reconstruction fidelity. We train our full model on 3D cell image data and explore design trade-offs in the 2D setting. Once trained, our model can be used to predict plausible locations of structures in cells where these structures were not imaged. The trained model can also be used to quantify the variation in the location of subcellular structures by generating plausible instantiations of each structure in arbitrary cell geometries. We apply our trained model to a small drug perturbation screen to demonstrate its applicability to new data. We show how the latent representations of drugged cells differ from unperturbed cells as expected by on-target effects of the drugs.

ER membrane contact sites: key platforms for biogenesis of RNA-containing extracellular vesicles.

The mechanisms by which cytoplasmic cargoes such as RNAs are incorporated into extracellular vesicles (EVs) are poorly understood. In a recent article published in Developmental Cell, we describe a novel function of endoplasmic reticulum membrane contact sites (ER MCS) in regulating biogenesis of RNA-containing EVs (Barman et al., 2022). We identified the ER MCS tether protein VAP-A and the ceramide transporter CERT as key drivers of this process. VAP-A depletion and overexpression produced corresponding changes in the overall number and RNA content of secreted EVs. Further sub-fractionation of small EVs from VAP-A depleted cells revealed a distinct loss in a specific subset of dense, RNA-loaded small EVs that are critical for the transfer of miR-100 to recipient cells. Cell imaging data confirmed the loss of RNA and RNA binding proteins (RBPs) in VAP-A-knockdown multivesicular bodies. Lipid analysis of VAP-A-knockdown EVs revealed decreases in ceramides, which are known to affect EV biogenesis. Depletion of the ceramide transfer protein CERT, which interacts with its binding partner VAP-A at ER MCS, leads to similar defects in EV number and RNA content as VAP-A-knockdown. These data suggest a model for ER MCS as platforms for biogenesis of a key EV population via ceramide transfer and RNA loading.

Correlation between Retinol Binding Protein 4 and CT Perfusion Imaging and Prognosis in Patients with Acute Ischemic Stroke Based on Cell Imaging Data Analysis.

Objective In order to explore the correlation between the retinol binding protein 4 (RBP4) and prognosis of patients with acute ischemic stroke (AIS), CT perfusion imaging can be used to scan the brain tissue of patients, which can identify abnormal perfusion areas and ischemic penumbra brain tissue, so as to provide a basis for doctors to formulate a reasonable clinical treatment plan. Methods 200 patients with first-episode acute ischemic stroke were selected from the Department of Neurology of our hospital, including 128 males and 72 females, aged between 32 and 85 years, with an average age of 45 ± 4.3 years. After admission, the patients were tested for xanthol binding protein 4 in time, the patient's demographic data and the basic clinical data were recorded, the degree of brain injury was evaluated, and the short-term outcome was evaluated after treatment. Results A total of 200 AIS patients with different degrees of brain injury were included in this study, including 128 males and 78 females, aged between 32 and 85 years, with an average age of 45 ± 4.3 years. Among them, 100 patients used CT perfusion imaging for brain scanning as the observation group, 100 patients used traditional imaging methods as the reference group, and 100 healthy people were included as the blank group. At the same time, the contents of total cholesterol, triacylglycerol, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol were measured by using an automatic biochemical analyzer. The patients were evaluated in the early stage of treatment and the effect of prognostic intervention was recorded. Conclusion The application of CT perfusion imaging in the adjuvant treatment of AIS patients was helpful to identify the abnormal perfusion area and the brain tissue of ischemic penumbra, so as to provide a basis for doctors' follow-up treatment. At the same time, AIS patients with high serum RBP4 level had mild stroke severity, good short-term prognosis, and improved treatment effect, which improved patients' quality of life.

A Study on the Detection of Defects in Solar Cells and Modules in EL Images based on Artificial Intelligence

Currently, investment is being made in renewable energy for the transition to a low-carbon economy and society, and interest in solar energy is also increasing. In addition to the technological development of solar cells and photovoltaic (PV) modules, research in the field of convergence with artificial intelligence technology is being actively conducted. Defects occurring in the manufacturing process of solar cells and modules can be detected through electroluminescence (EL) measurements. In this study, we propose an artificial intelligence technology that can automatically detect defects in cells and modules in real time using EL image data of solar cells and modules in the manufacturing process. For EL defect detection, we propose an algorithm with high suitability in terms of speed and accuracy by applying deep learning-based object detection models and comparing and evaluating detection performance. In the case of the YOLO (you only look once) algorithm, which belongs to a one-step detector, it learns In the case of the YOLO (you only look once) algorithm, which belongs to a one-step detector, it learns through an optimization process to find information about the defect and the location information of the corresponding failure in the form of a bounding box, and then detects the failure based on this information. The introduction of a deep learning-based defect detection model in the manufacturing process is expected to reduce the time required for defect detection by solar cell and PV module manufacturers and contribute to productivity improvement.

Automated Classification of Breast Cancer Cells Using High-Throughput Holographic Cytometry

Holographic cytometry is an ultra-high throughput quantitative phase imaging modality that is capable of extracting subcellular information from millions of cells flowing through parallel microfluidic channels. In this study, we present our findings on the application of holographic cytometry to distinguishing carcinogen-exposed cells from normal cells and cancer cells. This has potential application for environmental monitoring and cancer detection by analysis of cytology samples acquired via brushing or fine needle aspiration. By leveraging the vast amount of cell imaging data, we are able to build single-cell-analysis-based biophysical phenotype profiles on the examined cell lines. Multiple physical characteristics of these cells show observable distinct traits between the three cell types. Logistic regression analysis provides insight on which traits are more useful for classification. Additionally, we demonstrate that deep learning is a powerful tool that can potentially identify phenotypic differences from reconstructed single-cell images. The high classification accuracy levels show the platform’s potential in being developed into a diagnostic tool for abnormal cell screening.

TMOD-13. IDENTIFYING DRIVERS IN THE CONVERGING SYNTENIC REGIONS OF SPONTANEOUS CANINE AND PEDIATRIC HIGH-GRADE GLIOMA USING IMAGING BASED CRISPR-CAS9 ARRAY SCREEN

Abstract Gliomas occur in companion dogs at rates comparable to humans, with short-snouted breeds such as boxers being more susceptible than others. The natural progression of cancer in the immuno-competent host allows companion dogs diagnosed with sporadic glioma as an optimal model for preclinical testing of therapeutic approaches with human relevance, including immunotherapies. We have recently performed comprehensive genomic and epigenetic characterization of glioma in dogs to their human counterparts and found strong convergent evolution – shared somatic mutations and aneuploidies - among syntenic regions, including those of known pediatric glioma drivers, e.g., PDGFRA, MYC, PIK3CA. Here, using arrayed CRISPR-Cas9 imaging based phenotypic screen, we will probe potential oncogenic drivers and tumor suppressor genes within syntenic aneuploidies and thus outline functional versus non-functional heterogeneity of cancer aneuploidy. Specifically, we are conducting arrayed knockout screen (one gene per well) of 400+ genes within syntenic aneuploidies across canine (n=2) and pediatric (n=2) high-grade glioma cell lines. We will first capture images by high-speed confocal imaging system at three time points post-transduction of single guide RNAs (2 per gene) targeting each of 400+ genes in their separate wells. Then, using high-throughput image analysis and semi-supervised machine learning methods, we will measure well-based phenotypic features (viability, growth, and morphology) from these images. Genes will be ranked per cross-validated predicted probability in yielding either proliferating or slow-growing cell type based on learned phenotypic features using image data of knockout cells from and across wells. The top ranked genes will then be linked to oncogenes and tumor suppressors based on pathway and ontology analysis. We expect that we will see convergence of the most impactful molecular abnormalities (based on their knockout phenotypes) on mechanisms or candidate signaling pathways for the development of new drugs and repurposing of existing drugs for kids and dogs with high-grade glioma.

Highly Reversible Supramolecular Light Switch for NIR Phosphorescence Resonance Energy Transfer.

Although purely organic room‐temperature phosphorescence (RTP) has drawn widespread attention in recent years, regulatable phosphorescence resonance energy transfer (PRET) supramolecular switch is still rare. Herein, single molecular dual‐fold supramolecular light switches, which are constructed by phenylpyridinium salts modified diarylethene derivatives (DTE‐Cn, n = 3, 5) and cucurbit[8]uril (CB[8]) are reported. Significantly, biaxial [3]pseudorotaxane displayed efficiently reversible RTP after binding with CB[8] and the phosphorescence quenching efficiency is calculated up to be 99%. Furthermore, the binary supramolecular assembly can coassemble with Cy5 to form ternary supramolecular assembly showing efficiently PRET, which is successfully applied in switchable near infrared (NIR) mitochondria‐targeted cell imaging and photocontrolled data encryption. This supramolecular strategy involving energy transfer provides a convenient approach for phosphorescent application in biology and material fields.

Single Cell Analysis of Stored Red Blood Cells Using Ultra-High Throughput Holographic Cytometry

Holographic cytometry is introduced as an ultra-high throughput implementation of quantitative phase imaging of single cells flowing through parallel microfluidic channels. Here, the approach was applied for characterizing the morphology of individual red blood cells during storage under regular blood bank conditions. Samples from five blood donors were examined, over 100,000 cells examined for each, at three time points. The approach allows high-throughput phase imaging of a large number of cells, greatly extending our ability to study cellular phenotypes using individual cell images. Holographic cytology images can provide measurements of multiple physical traits of the cells, including optical volume and area, which are observed to consistently change over the storage time. In addition, the large volume of cell imaging data can serve as training data for machine-learning algorithms. For the study here, logistic regression was used to classify the cells according to the storage time points. The analysis showed that at least 5000 cells are needed to ensure accuracy of the classifiers. Overall, results showed the potential of holographic cytometry as a diagnostic tool.

Machine learning-assisted imaging analysis of a human epiblast model.

The human embryo is a complex structure that emerges and develops as a result of cell-level decisions guided by both intrinsic genetic programs and cell-cell interactions. Given limited accessibility and associated ethical constraints of human embryonic tissue samples, researchers have turned to the use of human stem cells to generate embryo models to study specific embryogenic developmental steps. However, to study complex self-organizing developmental events using embryo models, there is a need for computational and imaging tools for detailed characterization of cell-level dynamics at the single cell level. In this work, we obtained live cell imaging data from a human pluripotent stem cell (hPSC)-based epiblast model that can recapitulate the lumenal epiblast cyst formation soon after implantation of the human blastocyst. By processing imaging data with a Python pipeline that incorporates both cell tracking and event recognition with the use of a CNN-LSTM machine learning model, we obtained detailed temporal information of changes in cell state and neighborhood during the dynamic growth and morphogenesis of lumenal hPSC cysts. The use of this tool combined with reporter lines for cell types of interest will drive future mechanistic studies of hPSC fate specification in embryo models and will advance our understanding of how cell-level decisions lead to global organization and emergent phenomena. Insight, innovation, integration: Human pluripotent stem cells (hPSCs) have been successfully used to model and understand cellular events that take place during human embryogenesis. Understanding how cell-cell and cell-environment interactions guide cell actions within a hPSC-based embryo model is a key step in elucidating the mechanisms driving system-level embryonic patterning and growth. In this work, we present a robust video analysis pipeline that incorporates the use of machine learning methods to fully characterize the process of hPSC self-organization into lumenal cysts to mimic the lumenal epiblast cyst formation soon after implantation of the human blastocyst. This pipeline will be a useful tool for understanding cellular mechanisms underlying key embryogenic events in embryo models.

Crosstalk between astrocytes and microglia results in increased degradation of \u03b1-synuclein and amyloid-\u03b2 aggregates

BackgroundAlzheimer’s disease (AD) and Parkinson’s disease (PD) are characterized by brain accumulation of aggregated amyloid-beta (Aβ) and alpha-synuclein (αSYN), respectively. In order to develop effective therapies, it is crucial to understand how the Aβ/αSYN aggregates can be cleared. Compelling data indicate that neuroinflammatory cells, including astrocytes and microglia, play a central role in the pathogenesis of AD and PD. However, how the interplay between the two cell types affects their clearing capacity and consequently the disease progression remains unclear.MethodsThe aim of the present study was to investigate in which way glial crosstalk influences αSYN and Aβ pathology, focusing on accumulation and degradation. For this purpose, human-induced pluripotent cell (hiPSC)-derived astrocytes and microglia were exposed to sonicated fibrils of αSYN or Aβ and analyzed over time. The capacity of the two cell types to clear extracellular and intracellular protein aggregates when either cultured separately or in co-culture was studied using immunocytochemistry and ELISA. Moreover, the capacity of cells to interact with and process protein aggregates was tracked using time-lapse microscopy and a customized “close-culture” chamber, in which the apical surfaces of astrocyte and microglia monocultures were separated by a <1 mm space.ResultsOur data show that intracellular deposits of αSYN and Aβ are significantly reduced in co-cultures of astrocytes and microglia, compared to monocultures of either cell type. Analysis of conditioned medium and imaging data from the “close-culture” chamber experiments indicate that astrocytes secrete a high proportion of their internalized protein aggregates, while microglia do not. Moreover, co-cultured astrocytes and microglia are in constant contact with each other via tunneling nanotubes and other membrane structures. Notably, our live cell imaging data demonstrate that microglia, when attached to the cell membrane of an astrocyte, can attract and clear intracellular protein deposits from the astrocyte.ConclusionsTaken together, our data demonstrate the importance of astrocyte and microglia interactions in Aβ/αSYN clearance, highlighting the relevance of glial cellular crosstalk in the progression of AD- and PD-related brain pathology.

Interpretable deep learning uncovers cellular properties in label-free live cell images that are predictive of highly metastatic melanoma.

Deep learning has emerged as the technique of choice for identifying hidden patterns in cell imaging data but is often criticized as "black box." Here, we employ a generative neural network in combination with supervised machine learning to classify patient-derived melanoma xenografts as "efficient" or "inefficient" metastatic, validate predictions regarding melanoma cell lines with unknown metastatic efficiency in mouse xenografts, and use the network to generate in silico cell images that amplify the critical predictive cell properties. These exaggerated images unveiled pseudopodial extensions and increased light scattering as hallmark properties of metastatic cells. We validated this interpretation using live cells spontaneously transitioning between states indicative of low and high metastatic efficiency. This study illustrates how the application of artificial intelligence can support the identification of cellular properties that are predictive of complex phenotypes and integrated cell functions but are too subtle to be identified in the raw imagery by a human expert. A record of this paper's transparent peer review process is included in the supplemental information. VIDEO ABSTRACT.