Enhanced Cell Tracking Using A GAN-based Super-Resolution Video-to-Video Time-Lapse Microscopy Generative Model.

Cells are among the most dynamic entities, constantly undergoing various processes such as growth, division, movement, and interaction with other cells as well as the environment. Time-lapse microscopy is central to capturing these dynamic behaviors, providing detailed temporal and spatial information that allows biologists to observe and analyze cellular activities in real-time. The analysis of time-lapse microscopy data relies on two fundamental tasks: cell segmentation and cell tracking. Integrating deep learning into bioimage analysis has revolutionized cell segmentation, producing models with high precision across a wide range of biological images. However, developing generalizable deep-learning models for tracking cells over time remains challenging due to the scarcity of large, diverse annotated datasets of time-lapse movies of cells. To address this bottleneck, we propose a GAN-based time-lapse microscopy generator, termed tGAN, designed to significantly enhance the quality and diversity of synthetic annotated time-lapse microscopy data. Our model features a dual-resolution architecture that adeptly synthesizes both low and high-resolution images, uniquely capturing the intricate dynamics of cellular processes essential for accurate tracking. We demonstrate the performance of tGAN in generating high-quality, realistic, annotated time-lapse videos. Our findings indicate that tGAN decreases dependency on extensive manual annotation to enhance the precision of cell tracking models for time-lapse microscopy.

Enhanced cell segmentation with limited training datasets using cycle generative adversarial networks.

Deep learning is transforming bioimage analysis, but its application in single-cell segmentation is limited by the lack of large, diverse annotated datasets. We addressed this by introducing a CycleGAN-based architecture, cGAN-Seg, that enhances the training of cell segmentation models with limited annotated datasets. During training, cGAN-Seg generates annotated synthetic phase-contrast or fluorescent images with morphological details and nuances closely mimicking real images. This increases the variability seen by the segmentation model, enhancing the authenticity of synthetic samples and thereby improving predictive accuracy and generalization. Experimental results show that cGAN-Seg significantly improves the performance of widely used segmentation models over conventional training techniques. Our approach has the potential to accelerate the development of foundation models for microscopy image analysis, indicating its significance in advancing bioimage analysis with efficient training methodologies.

Self-organization of embryonic stem cells into a reproducible embryo model through epigenome editing.

Embryonic stem cells (ESCs) can self-organize in vitro into developmental patterns with spatial organization and molecular similarity to that of early embryonic stages. This self-organization of ESCs requires transmission of signaling cues, via addition of small molecule chemicals or recombinant proteins, to induce distinct embryonic cellular fates and subsequent assembly into structures that can mimic aspects of early embryonic development. During natural embryonic development, different embryonic cell types co-develop together, where each cell type expresses specific fate-inducing transcription factors through activation of non-coding regulatory elements and interactions with neighboring cells. However, previous studies have not fully explored the possibility of engineering endogenous regulatory elements to shape self-organization of ESCs into spatially-ordered embryo models. Here, we hypothesized that cell-intrinsic activation of a minimum number of such endogenous regulatory elements is sufficient to self-organize ESCs into early embryonic models. Our results show that CRISPR-based activation (CRISPRa) of only two endogenous regulatory elements in the genome of pluripotent stem cells is sufficient to generate embryonic patterns that show spatial and molecular resemblance to that of pre-gastrulation mouse embryonic development. Quantitative single-cell live fluorescent imaging showed that the emergence of spatially-ordered embryonic patterns happens through the intrinsic induction of cell fate that leads to an orchestrated collective cellular motion. Based on these results, we propose a straightforward approach to efficiently form 3D embryo models through intrinsic CRISPRa-based epigenome editing and independent of external signaling cues. CRISPRa-Programmed Embryo Models (CPEMs) show highly consistent composition of major embryonic cell types that are spatially-organized, with nearly 80% of the structures forming an embryonic cavity. Single cell transcriptomics confirmed the presence of main embryonic cell types in CPEMs with transcriptional similarity to pre-gastrulation mouse embryos and revealed novel signaling communication links between different embryonic cell types. Our findings offer a programmable embryo model and demonstrate that minimum intrinsic epigenome editing is sufficient to self-organize ESCs into highly consistent pre-gastrulation embryo models.

Enhanced cell segmentation with limited annotated data using generative adversarial networks.

The application of deep learning is rapidly transforming the field of bioimage analysis. While deep learning has shown great promise in complex microscopy tasks such as single-cell segmentation, the development of generalizable foundation deep learning segmentation models is hampered by the scarcity of large and diverse annotated datasets of cell images for training purposes. Generative Adversarial Networks (GANs) can generate realistic images that can potentially be easily used to train deep learning models without the generation of large manually annotated microscopy images. Here, we propose a customized CycleGAN architecture to train an enhanced cell segmentation model with limited annotated cell images, effectively addressing the challenge of paucity of annotated data in microscopy imaging. Our customized CycleGAN model can generate realistic synthetic images of cells with morphological details and nuances very similar to that of real images. This method not only increases the variability seen during training but also enhances the authenticity of synthetic samples, thereby enhancing the overall predictive accuracy and robustness of the cell segmentation model. Our experimental results show that our CycleGAN-based method significantly improves the performance of the segmentation model compared to conventional training techniques. Interestingly, we demonstrate that our model can extrapolate its knowledge by synthesizing imaging scenarios that were not seen during the training process. Our proposed customized CycleGAN method will accelerate the development of foundation models for cell segmentation in microscopy images.

DeepSea is an efficient deep-learning model for single-cell segmentation and tracking in time-lapse microscopy.

Time-lapse microscopy is the only method that can directly capture the dynamics and heterogeneity of fundamental cellular processes at the single-cell level with high temporal resolution. Successful application of single-cell time-lapse microscopy requires automated segmentation and tracking of hundreds of individual cells over several time points. However, segmentation and tracking of single cells remain challenging for the analysis of time-lapse microscopy images, in particular for widely available and non-toxic imaging modalities such as phase-contrast imaging. This work presents a versatile and trainable deep-learning model, termed DeepSea, that allows for both segmentation and tracking of single cells in sequences of phase-contrast live microscopy images with higher precision than existing models. We showcase the application of DeepSea by analyzing cell size regulation in embryonic stem cells.

COVID-Classifier: an automated machine learning model to assist in the diagnosis of COVID-19 infection in chest X-ray images.

Chest-X ray (CXR) radiography can be used as a first-line triage process for non-COVID-19 patients with pneumonia. However, the similarity between features of CXR images of COVID-19 and pneumonia caused by other infections makes the differential diagnosis by radiologists challenging. We hypothesized that machine learning-based classifiers can reliably distinguish the CXR images of COVID-19 patients from other forms of pneumonia. We used a dimensionality reduction method to generate a set of optimal features of CXR images to build an efficient machine learning classifier that can distinguish COVID-19 cases from non-COVID-19 cases with high accuracy and sensitivity. By using global features of the whole CXR images, we successfully implemented our classifier using a relatively small dataset of CXR images. We propose that our COVID-Classifier can be used in conjunction with other tests for optimal allocation of hospital resources by rapid triage of non-COVID-19 cases.

COVID-Classifier: An automated machine learning model to assist in the diagnosis of COVID-19 infection in chest x-ray images.

Chest-X ray (CXR) radiography can be used as a first-line triage process for non-COVID-19 patients with pneumonia. However, the similarity between features of CXR images of COVID-19 and pneumonia caused by other infections make the differential diagnosis by radiologists challenging. We hypothesized that machine learning-based classifiers can reliably distinguish the CXR images of COVID-19 patients from other forms of pneumonia. We used a dimensionality reduction method to generate a set of optimal features of CXR images to build an efficient machine learning classifier that can distinguish COVID-19 cases from non-COVID-19 cases with high accuracy and sensitivity. By using global features of the whole CXR images, we were able to successfully implement our classifier using a relatively small dataset of CXR images. We propose that our COVID-Classifier can be used in conjunction with other tests for optimal allocation of hospital resources by rapid triage of non-COVID-19 cases.

Microtechnology-based methods for organoid models.

Innovations in biomaterials and stem cell technology have allowed for the emergence of novel three-dimensional (3D) tissue-like structures known as organoids and spheroids. As a result, compared to conventional 2D cell culture and animal models, these complex 3D structures have improved the accuracy and facilitated in vitro investigations of human diseases, human development, and personalized medical treatment. Due to the rapid progress of this field, numerous spheroid and organoid production methodologies have been published. However, many of the current spheroid and organoid production techniques are limited by complexity, throughput, and reproducibility. Microfabricated and microscale platforms (e.g., microfluidics and microprinting) have shown promise to address some of the current limitations in both organoid and spheroid generation. Microfabricated and microfluidic devices have been shown to improve nutrient delivery and exchange and have allowed for the arrayed production of size-controlled culture areas that yield more uniform organoids and spheroids for a higher throughput at a lower cost. In this review, we discuss the most recent production methods, challenges currently faced in organoid and spheroid production, and microfabricated and microfluidic applications for improving spheroid and organoid generation. Specifically, we focus on how microfabrication methods and devices such as lithography, microcontact printing, and microfluidic delivery systems can advance organoid and spheroid applications in medicine.

Reversible Disruption of Specific Transcription Factor-DNA Interactions Using CRISPR/Cas9.

The control of gene expression by transcription factor binding sites frequently determines phenotype. However, it is difficult to determine the function of single transcription factor binding sites within larger transcription networks. Here, we use deactivated Cas9 (dCas9) to disrupt binding to specific sites, a method we term CRISPRd. Since CRISPR guide RNAs are longer than transcription factor binding sites, flanking sequence can be used to target specific sites. Targeting dCas9 to an Oct4 site in the Nanog promoter displaced Oct4 from this site, reduced Nanog expression, and slowed division. In contrast, disrupting the Oct4 binding site adjacent to Pax6 upregulated Pax6 transcription and disrupting Nanog binding its own promoter upregulated its transcription. Thus, we can easily distinguish between activating and repressing binding sites and examine autoregulation. Finally, multiple guide RNA expression allows simultaneous inhibition of multiple binding sites, and conditionally destabilized dCas9 allows rapid reversibility.

Zygotic Genome Activation in Vertebrates.

The first major developmental transition in vertebrate embryos is the maternal-to-zygotic transition (MZT) when maternal mRNAs are degraded and zygotic transcription begins. During the MZT, the embryo takes charge of gene expression to control cell differentiation and further development. This spectacular organismal transition requires nuclear reprogramming and the initiation of RNAPII at thousands of promoters. Zygotic genome activation (ZGA) is mechanistically coordinated with other embryonic events, including changes in the cell cycle, chromatin state, and nuclear-to-cytoplasmic component ratios. Here, we review progress in understanding vertebrate ZGA dynamics in frogs, fish, mice, and humans to explore differences and emphasize common features.

Cell autonomous regulation of hippocampal circuitry via Aph1b-γ-secretase/neuregulin 1 signalling.

Neuregulin 1 (NRG1) and the γ-secretase subunit APH1B have been previously implicated as genetic risk factors for schizophrenia and schizophrenia relevant deficits have been observed in rodent models with loss of function mutations in either gene. Here we show that the Aph1b-γ-secretase is selectively involved in Nrg1 intracellular signalling. We found that Aph1b-deficient mice display a decrease in excitatory synaptic markers. Electrophysiological recordings show that Aph1b is required for excitatory synaptic transmission and plasticity. Furthermore, gain and loss of function and genetic rescue experiments indicate that Nrg1 intracellular signalling promotes dendritic spine formation downstream of Aph1b-γ-secretase in vitro and in vivo. In conclusion, our study sheds light on the physiological role of Aph1b-γ-secretase in brain and provides a new mechanistic perspective on the relevance of NRG1 processing in schizophrenia.

Redundancy and divergence in the amyloid precursor protein family.

Gene duplication provides genetic material required for functional diversification. An interesting example is the amyloid precursor protein (APP) protein family. The APP gene family has experienced both expansion and contraction during evolution. The three mammalian members have been studied quite extensively in combined knock out models. The underlying assumption is that APP, amyloid precursor like protein 1 and 2 (APLP1, APLP2) are functionally redundant. This assumption is primarily supported by the similarities in biochemical processing of APP and APLPs and on the fact that the different APP genes appear to genetically interact at the level of the phenotype in combined knockout mice. However, unique features in each member of the APP family possibly contribute to specification of their function. In the current review, we discuss the evolution and the biology of the APP protein family with special attention to the distinct properties of each homologue. We propose that the functions of APP, APLP1 and APLP2 have diverged after duplication to contribute distinctly to different neuronal events. Our analysis reveals that APLP2 is significantly diverged from APP and APLP1.

APLP2 regulates neuronal stem cell differentiation during cortical development.

Expression of amyloid precursor protein (APP) and its two paralogues, APLP1 and APLP2 during brain development coincides with key cellular events such as neuronal differentiation and migration. However, genetic knockout and shRNA studies have led to contradictory conclusions about their role during embryonic brain development. To address this issue, we analysed in depth the role of APLP2 during neurogenesis by silencing APLP2 in vivo in an APP/APLP1 double knockout mouse background. We find that under these conditions cortical progenitors remain in their undifferentiated state much longer, displaying a higher number of mitotic cells. In addition, we show that neuron-specific APLP2 downregulation does not impact the speed or position of migrating excitatory cortical neurons. In summary, our data reveal that APLP2 is specifically required for proper cell cycle exit of neuronal progenitors, and thus has a distinct role in priming cortical progenitors for neuronal differentiation.

Analysis of microRNA signatures using size-coded ligation-mediated PCR.

The expression pattern and regulatory functions of microRNAs (miRNAs) are intensively investigated in various tissues, cell types and disorders. Differential miRNA expression signatures have been revealed in healthy and unhealthy tissues using high-throughput profiling methods. For further analyses of miRNA signatures in biological samples, we describe here a simple and efficient method to detect multiple miRNAs simultaneously in total RNA. The size-coded ligation-mediated polymerase chain reaction (SL-PCR) method is based on size-coded DNA probe hybridization in solution, followed-by ligation, PCR amplification and gel fractionation. The new method shows quantitative and specific detection of miRNAs. We profiled miRNAs of the let-7 family in a number of organisms, tissues and cell types and the results correspond with their incidence in the genome and reported expression levels. Finally, SL-PCR detected let-7 expression changes in human embryonic stem cells as they differentiate to neuron and also in young and aged mice brain and bone marrow. We conclude that the method can efficiently reveal miRNA signatures in a range of biological samples.

Neurons generated from APP/APLP1/APLP2 triple knockout embryonic stem cells behave normally in vitro and in vivo: lack of evidence for a cell autonomous role of the amyloid precursor protein in neuronal differentiation.

Alzheimer's disease amyloid precursor protein (APP) has been implicated in many neurobiologic processes, but supporting evidence remains indirect. Studies are confounded by the existence of two partially redundant APP homologues, APLP1 and APLP2. APP/APLP1/APLP2 triple knockout (APP tKO) mice display cobblestone lissencephaly and are perinatally lethal. To circumvent this problem, we generated APP triple knockout embryonic stem (ES) cells and differentiated these to APP triple knockout neurons in vitro and in vivo. In comparison with wild-type (WT) ES cell-derived neurons, APP tKO neurons formed equally pure neuronal cultures, had unaltered in vitro migratory capacities, had a similar acquisition of polarity, and were capable of extending long neurites and forming active excitatory synapses. These data were confirmed in vivo in chimeric mice with APP tKO neurons expressing the enhanced green fluorescent protein (eGFP) present in a WT background brain. The results suggest that the loss of the APP family of proteins has no major effect on these critical neuronal processes and that the apparent multitude of functions in which APP has been implicated might be characterized by molecular redundancy. Our stem cell culture provides an excellent tool to circumvent the problem of lack of viability of APP/APLP triple knockout mice and will help to explore the function of this intriguing protein further in vitro and in vivo.