ReCSAI: recursive compressed sensing artificial intelligence for confocal lifetime localization microscopy.

BACKGROUND: Localization-based super-resolution microscopy resolves macromolecular structures down to a few nanometers by computationally reconstructing fluorescent emitter coordinates from diffraction-limited spots. The most commonly used algorithms are based on fitting parametric models of the point spread function (PSF) to a measured photon distribution. These algorithms make assumptions about the symmetry of the PSF and thus, do not work well with irregular, non-linear PSFs that occur for example in confocal lifetime imaging, where a laser is scanned across the sample. An alternative method for reconstructing sparse emitter sets from noisy, diffraction-limited images is compressed sensing, but due to its high computational cost it has not yet been widely adopted. Deep neural network fitters have recently emerged as a new competitive method for localization microscopy. They can learn to fit arbitrary PSFs, but require extensive simulated training data and do not generalize well. A method to efficiently fit the irregular PSFs from confocal lifetime localization microscopy combining the advantages of deep learning and compressed sensing would greatly improve the acquisition speed and throughput of this method. RESULTS: Here we introduce ReCSAI, a compressed sensing neural network to reconstruct localizations for confocal dSTORM, together with a simulation tool to generate training data. We implemented and compared different artificial network architectures, aiming to combine the advantages of compressed sensing and deep learning. We found that a U-Net with a recursive structure inspired by iterative compressed sensing showed the best results on realistic simulated datasets with noise, as well as on real experimentally measured confocal lifetime scanning data. Adding a trainable wavelet denoising layer as prior step further improved the reconstruction quality. CONCLUSIONS: Our deep learning approach can reach a similar reconstruction accuracy for confocal dSTORM as frame binning with traditional fitting without requiring the acquisition of multiple frames. In addition, our work offers generic insights on the reconstruction of sparse measurements from noisy experimental data by combining compressed sensing and deep learning. We provide the trained networks, the code for network training and inference as well as the simulation tool as python code and Jupyter notebooks for easy reproducibility.

Biological network growth in complex environments: A computational framework.

Spatial biological networks are abundant on all scales of life, from single cells to ecosystems, and perform various important functions including signal transmission and nutrient transport. These biological functions depend on the architecture of the network, which emerges as the result of a dynamic, feedback-driven developmental process. While cell behavior during growth can be genetically encoded, the resulting network structure depends on spatial constraints and tissue architecture. Since network growth is often difficult to observe experimentally, computer simulations can help to understand how local cell behavior determines the resulting network architecture. We present here a computational framework based on directional statistics to model network formation in space and time under arbitrary spatial constraints. Growth is described as a biased correlated random walk where direction and branching depend on the local environmental conditions and constraints, which are presented as 3D multilayer grid. To demonstrate the application of our tool, we perform growth simulations of a dense network between cells and compare the results to experimental data from osteocyte networks in bone. Our generic framework might help to better understand how network patterns depend on spatial constraints, or to identify the biological cause of deviations from healthy network function.

Haruspex: A Neural Network for the Automatic Identification of Oligonucleotides and Protein Secondary Structure in Cryo-Electron Microscopy Maps.

In recent years, three-dimensional density maps reconstructed from single particle images obtained by electron cryo-microscopy (cryo-EM) have reached unprecedented resolution. However, map interpretation can be challenging, in particular if the constituting structures require de-novo model building or are very mobile. Herein, we demonstrate the potential of convolutional neural networks for the annotation of cryo-EM maps: our network Haruspex has been trained on a carefully curated set of 293 experimentally derived reconstruction maps to automatically annotate RNA/DNA as well as protein secondary structure elements. It can be straightforwardly applied to newly reconstructed maps in order to support domain placement or as a starting point for main-chain placement. Due to its high recall and precision rates of 95.1 % and 80.3 %, respectively, on an independent test set of 122 maps, it can also be used for validation during model building. The trained network will be available as part of the CCP-EM suite.

Towards a Connectomic Description of the Osteocyte Lacunocanalicular Network in Bone.

PURPOSE OF REVIEW: Osteocytes are the most abundant bone cells. They are completely encased in mineralized tissue, sitting inside lacunae that are connected by a multitude of canaliculi. In recent years, the osteocyte network has been shown to fulfill endocrine functions and to communicate with a number of other organs. This review addresses emerging knowledge on the connectome of the lacunocanalicular network in different types of bone tissue. RECENT FINDINGS: Recent advances in three-dimensional imaging technology started to reveal parameters that are well known from general theory to characterize the function of networks, such as network density, degree of nodes, or shortest path length through the network. The connectome of the lacunocanalicular network differs in some aspects between lamellar and woven bone and seems to change with age. More research is needed to relate network structure to function, such as intercellular transport or communication and its role in mechanosensation, as well as to understand the effect of diseases.

FIJI Macro 3D ART VeSElecT: 3D Automated Reconstruction Tool for Vesicle Structures of Electron Tomograms.

Automatic image reconstruction is critical to cope with steadily increasing data from advanced microscopy. We describe here the Fiji macro 3D ART VeSElecT which we developed to study synaptic vesicles in electron tomograms. We apply this tool to quantify vesicle properties (i) in embryonic Danio rerio 4 and 8 days past fertilization (dpf) and (ii) to compare Caenorhabditis elegans N2 neuromuscular junctions (NMJ) wild-type and its septin mutant (unc-59(e261)). We demonstrate development-specific and mutant-specific changes in synaptic vesicle pools in both models. We confirm the functionality of our macro by applying our 3D ART VeSElecT on zebrafish NMJ showing smaller vesicles in 8 dpf embryos then 4 dpf, which was validated by manual reconstruction of the vesicle pool. Furthermore, we analyze the impact of C. elegans septin mutant unc-59(e261) on vesicle pool formation and vesicle size. Automated vesicle registration and characterization was implemented in Fiji as two macros (registration and measurement). This flexible arrangement allows in particular reducing false positives by an optional manual revision step. Preprocessing and contrast enhancement work on image-stacks of 1nm/pixel in x and y direction. Semi-automated cell selection was integrated. 3D ART VeSElecT removes interfering components, detects vesicles by 3D segmentation and calculates vesicle volume and diameter (spherical approximation, inner/outer diameter). Results are collected in color using the RoiManager plugin including the possibility of manual removal of non-matching confounder vesicles. Detailed evaluation considered performance (detected vesicles) and specificity (true vesicles) as well as precision and recall. We furthermore show gain in segmentation and morphological filtering compared to learning based methods and a large time gain compared to manual segmentation. 3D ART VeSElecT shows small error rates and its speed gain can be up to 68 times faster in comparison to manual annotation. Both automatic and semi-automatic modes are explained including a tutorial.

Coalignment of osteocyte canaliculi and collagen fibers in human osteonal bone.

During bone formation osteocytes get connected with each other via a dense network of canaliculi within the mineralized bone matrix. Important functions attributed to the osteocyte network include the control of bone remodeling and a contribution to mineral homeostasis. To detect structural clues of the formation and functionality of the network, this study analyzes the structure and orientation of the osteocyte lacuno-canalicular network (OLCN), specifically in relation to the concentric bone lamellae within human osteons. The network structure within 49 osteons from four samples of cortical bone from the femoral midshaft of middle-aged healthy women was determined by a combination of rhodamine staining and confocal laser scanning microscopy followed by computational image analysis. A quantitative evaluation showed that 64±1% of the canalicular length has an angle smaller than 30° to the direction towards the osteon center, while the lateral network - defined by an orientation angle larger than 60° - comprises 16±1%. With the same spatial periodicity as the bone lamellae, both radial and lateral network show variations in the network density and order. However, only the preferred orientation of the lateral network twists when crossing a lamella. This twist agrees with the preferred orientation of the fibrous collagen matrix. The chirality of the twist was found to be individual-specific. The coalignment between network and matrix extends to the orientation of the elongated osteocyte lacunae. The intimate link between OLCN and collagen matrix implies an interplay between osteocyte processes and the arrangement of the surrounding collagen fibers during osteoid formation.

Fiber-assisted molding (FAM) of surfaces with tunable curvature to guide cell alignment and complex tissue architecture.

A simple and robust method termed "fiber-assisted molding (FAM)" is presented to create biomimetic three-dimensional surfaces with controllable curvature and helical twist. The alignment of muscle fibrils and the assembly of helically patterned extracellular matrix by cells demonstrate the potential of this method for tissue engineering and other materials science applications.

Tissues as networks of cells: towards generative rules of complex organ development.

Network analysis is a well-known and powerful tool in molecular biology. More recently, it has been introduced in developmental biology. Tissues can be readily translated into spatial networks such that cells are represented by nodes and intercellular connections by edges. This discretization of cellular organization enables mathematical approaches rooted in network science to be applied towards the understanding of tissue structure and function. Here, we describe how such tissue abstractions can enable the principles that underpin tissue formation and function to be uncovered. We provide an introduction into biologically relevant network measures, then present an overview of different areas of developmental biology where these approaches have been applied. We then summarize the general developmental rules underpinning tissue topology generation. Finally, we discuss how generative models can help to link the developmental rule back to the tissue topologies. Our collection of results points at general mechanisms as to how local developmental rules can give rise to observed topological properties in multicellular systems.

Nanoscaled RIM clustering at presynaptic active zones revealed by endogenous tagging.

Chemical synaptic transmission involves neurotransmitter release from presynaptic active zones (AZs). The AZ protein Rab-3-interacting molecule (RIM) is important for normal Ca2+-triggered release. However, its precise localization within AZs of the glutamatergic neuromuscular junctions of Drosophila melanogaster remains elusive. We used CRISPR/Cas9-assisted genome engineering of the rim locus to incorporate small epitope tags for targeted super-resolution imaging. A V5-tag, derived from simian virus 5, and an HA-tag, derived from human influenza virus, were N-terminally fused to the RIM Zinc finger. Whereas both variants are expressed in co-localization with the core AZ scaffold Bruchpilot, electrophysiological characterization reveals that AP-evoked synaptic release is disturbed in rimV5-Znf but not in rimHA-Znf In addition, rimHA-Znf synapses show intact presynaptic homeostatic potentiation. Combining super-resolution localization microscopy and hierarchical clustering, we detect ∼10 RIMHA-Znf subclusters with ∼13 nm diameter per AZ that are compacted and increased in numbers in presynaptic homeostatic potentiation.

Multicellular dynamics on structured surfaces: Stress concentration is a key to controlling complex microtissue morphology on engineered scaffolds.

Tissue engineers have utilised a variety of three-dimensional (3D) scaffolds for controlling multicellular dynamics and the resulting tissue microstructures. In particular, cutting-edge microfabrication technologies, such as 3D bioprinting, provide increasingly complex structures. However, unpredictable microtissue detachment from scaffolds, which ruins desired tissue structures, is becoming an evident problem. To overcome this issue, we elucidated the mechanism underlying collective cellular detachment by combining a new computational simulation method with quantitative tissue-culture experiments. We first quantified the stochastic processes of cellular detachment shown by vascular smooth muscle cells on model curved scaffolds and found that microtissue morphologies vary drastically depending on cell contractility, substrate curvature, and cell-substrate adhesion strength. To explore this mechanism, we developed a new particle-based model that explicitly describes stochastic processes of multicellular dynamics, such as adhesion, rupture, and large deformation of microtissues on structured surfaces. Computational simulations using the developed model successfully reproduced characteristic detachment processes observed in experiments. Crucially, simulations revealed that cellular contractility-induced stress is locally concentrated at the cell-substrate interface, subsequently inducing a catastrophic process of collective cellular detachment, which can be suppressed by modulating cell contractility, substrate curvature, and cell-substrate adhesion. These results show that the developed computational method is useful for predicting engineered tissue dynamics as a platform for prediction-guided scaffold design. STATEMENT OF SIGNIFICANCE: Microfabrication technologies aiming to control multicellular dynamics by engineering 3D scaffolds are attracting increasing attention for modelling in cell biology and regenerative medicine. However, obtaining microtissues with the desired 3D structures is made considerably more difficult by microtissue detachments from scaffolds. This study reveals a key mechanism behind this detachment by developing a novel computational method for simulating multicellular dynamics on designed scaffolds. This method enabled us to predict microtissue dynamics on structured surfaces, based on cell mechanics, substrate geometry, and cell-substrate interaction. This study provides a platform for the physics-based design of micro-engineered scaffolds and thus contributes to prediction-guided biomaterials design in the future.

How the mechanobiology orchestrates the iterative and reciprocal ECM-cell cross-talk that drives microtissue growth.

Controlled tissue growth is essential for multicellular life and requires tight spatiotemporal control over cell proliferation and differentiation until reaching homeostasis. As cells synthesize and remodel extracellular matrix, tissue growth processes can only be understood if the reciprocal feedback between cells and their environment is revealed. Using de novo-grown microtissues, we identified crucial actors of the mechanoregulated events, which iteratively orchestrate a sharp transition from tissue growth to maturation, requiring a myofibroblast-to-fibroblast transition. Cellular decision-making occurs when fibronectin fiber tension switches from highly stretched to relaxed, and it requires the transiently up-regulated appearance of tenascin-C and tissue transglutaminase, matrix metalloprotease activity, as well as a switch from α5ß1 to α2ß1 integrin engagement and epidermal growth factor receptor signaling. As myofibroblasts are associated with wound healing and inflammatory or fibrotic diseases, crucial knowledge needed to advance regenerative strategies or to counter fibrosis and cancer progression has been gained.

Predicting Microenvironment in CXCR4- and FAP-Positive Solid Tumors-A Pan-Cancer Machine Learning Workflow for Theranostic Target Structures.

(1) Background: C-X-C Motif Chemokine Receptor 4 (CXCR4) and Fibroblast Activation Protein Alpha (FAP) are promising theranostic targets. However, it is unclear whether CXCR4 and FAP positivity mark distinct microenvironments, especially in solid tumors. (2) Methods: Using Random Forest (RF) analysis, we searched for entity-independent mRNA and microRNA signatures related to CXCR4 and FAP overexpression in our pan-cancer cohort from The Cancer Genome Atlas (TCGA) database-representing n = 9242 specimens from 29 tumor entities. CXCR4- and FAP-positive samples were assessed via StringDB cluster analysis, EnrichR, Metascape, and Gene Set Enrichment Analysis (GSEA). Findings were validated via correlation analyses in n = 1541 tumor samples. TIMER2.0 analyzed the association of CXCR4 / FAP expression and infiltration levels of immune-related cells. (3) Results: We identified entity-independent CXCR4 and FAP gene signatures representative for the majority of solid cancers. While CXCR4 positivity marked an immune-related microenvironment, FAP overexpression highlighted an angiogenesis-associated niche. TIMER2.0 analysis confirmed characteristic infiltration levels of CD8+ cells for CXCR4-positive tumors and endothelial cells for FAP-positive tumors. (4) Conclusions: CXCR4- and FAP-directed PET imaging could provide a non-invasive decision aid for entity-agnostic treatment of microenvironment in solid malignancies. Moreover, this machine learning workflow can easily be transferred towards other theranostic targets.

Curvature in Biological Systems: Its Quantification, Emergence, and Implications across the Scales.

Surface curvature both emerges from, and influences the behavior of, living objects at length scales ranging from cell membranes to single cells to tissues and organs. The relevance of surface curvature in biology is supported by numerous experimental and theoretical investigations in recent years. In this review, first, a brief introduction to the key ideas of surface curvature in the context of biological systems is given and the challenges that arise when measuring surface curvature are discussed. Giving an overview of the emergence of curvature in biological systems, its significance at different length scales becomes apparent. On the other hand, summarizing current findings also shows that both single cells and entire cell sheets, tissues or organisms respond to curvature by modulating their shape and their migration behavior. Finally, the interplay between the distribution of morphogens or micro-organisms and the emergence of curvature across length scales is addressed with examples demonstrating these key mechanistic principles of morphogenesis. Overall, this review highlights that curved interfaces are not merely a passive by-product of the chemical, biological, and mechanical processes but that curvature acts also as a signal that co-determines these processes.

Detecting ice artefacts in processed macromolecular diffraction data with machine learning.

Contamination with diffraction from ice crystals can negatively affect, or even impede, macromolecular structure determination, and therefore detecting the resulting artefacts in diffraction data is crucial. However, once the data have been processed it can be very difficult to automatically recognize this problem. To address this, a set of convolutional neural networks named Helcaraxe has been developed which can detect ice-diffraction artefacts in processed diffraction data from macromolecular crystals. The networks outperform previous algorithms and will be available as part of the AUSPEX web server and the CCP4-distributed software.

Visual Clustering of Transcriptomic Data from Primary and Metastatic Tumors-Dependencies and Novel Pitfalls.

Personalized oncology is a rapidly evolving area and offers cancer patients therapy options that are more specific than ever. However, there is still a lack of understanding regarding transcriptomic similarities or differences of metastases and corresponding primary sites. Applying two unsupervised dimension reduction methods (t-Distributed Stochastic Neighbor Embedding (t-SNE) and Uniform Manifold Approximation and Projection (UMAP)) on three datasets of metastases (n =682 samples) with three different data transformations (unprocessed, log10 as well as log10 + 1 transformed values), we visualized potential underlying clusters. Additionally, we analyzed two datasets (n =616 samples) containing metastases and primary tumors of one entity, to point out potential familiarities. Using these methods, no tight link between the site of resection and cluster formation outcome could be demonstrated, or for datasets consisting of solely metastasis or mixed datasets. Instead, dimension reduction methods and data transformation significantly impacted visual clustering results. Our findings strongly suggest data transformation to be considered as another key element in the interpretation of visual clustering approaches along with initialization and different parameters. Furthermore, the results highlight the need for a more thorough examination of parameters used in the analysis of clusters.

Endogenous tagging of Unc-13 reveals nanoscale reorganization at active zones during presynaptic homeostatic potentiation.

Introduction: Neurotransmitter release at presynaptic active zones (AZs) requires concerted protein interactions within a dense 3D nano-hemisphere. Among the complex protein meshwork the (M)unc-13 family member Unc-13 of Drosophila melanogaster is essential for docking of synaptic vesicles and transmitter release. Methods: We employ minos-mediated integration cassette (MiMIC)-based gene editing using GFSTF (EGFP-FlAsH-StrepII-TEV-3xFlag) to endogenously tag all annotated Drosophila Unc-13 isoforms enabling visualization of endogenous Unc-13 expression within the central and peripheral nervous system. Results and discussion: Electrophysiological characterization using two-electrode voltage clamp (TEVC) reveals that evoked and spontaneous synaptic transmission remain unaffected in unc-13 GFSTF 3rd instar larvae and acute presynaptic homeostatic potentiation (PHP) can be induced at control levels. Furthermore, multi-color structured-illumination shows precise co-localization of Unc-13GFSTF, Bruchpilot, and GluRIIA-receptor subunits within the synaptic mesoscale. Localization microscopy in combination with HDBSCAN algorithms detect Unc-13GFSTF subclusters that move toward the AZ center during PHP with unaltered Unc-13GFSTF protein levels.

Ultrastructural analysis of wild-type and RIM1α knockout active zones in a large cortical synapse.

Rab3A-interacting molecule (RIM) is crucial for fast Ca2+-triggered synaptic vesicle (SV) release in presynaptic active zones (AZs). We investigated hippocampal giant mossy fiber bouton (MFB) AZ architecture in 3D using electron tomography of rapid cryo-immobilized acute brain slices in RIM1α-/- and wild-type mice. In RIM1α-/-, AZs are larger with increased synaptic cleft widths and a 3-fold reduced number of tightly docked SVs (0-2 nm). The distance of tightly docked SVs to the AZ center is increased from 110 to 195 nm, and the width of their electron-dense material between outer SV membrane and AZ membrane is reduced. Furthermore, the SV pool in RIM1α-/- is more heterogeneous. Thus, RIM1α, besides its role in tight SV docking, is crucial for synaptic architecture and vesicle pool organization in MFBs.

Vinculin facilitates cell invasion into three-dimensional collagen matrices.

The cytoskeletal protein vinculin contributes to the mechanical link of the contractile actomyosin cytoskeleton to the extracellular matrix (ECM) through integrin receptors. In addition, vinculin modulates the dynamics of cell adhesions and is associated with decreased cell motility on two-dimensional ECM substrates. The effect of vinculin on cell invasion through dense three-dimensional ECM gels is unknown. Here, we report how vinculin expression affects cell invasion into three-dimensional collagen matrices. Cell motility was investigated in vinculin knockout and vinculin expressing wild-type mouse embryonic fibroblasts. Vinculin knockout cells were 2-fold more motile on two-dimensional collagen-coated substrates compared with wild-type cells, but 3-fold less invasive in 2.4 mg/ml three-dimensional collagen matrices. Vinculin knockout cells were softer and remodeled their cytoskeleton more dynamically, which is consistent with their enhanced two-dimensional motility but does not explain their reduced three-dimensional invasiveness. Importantly, vinculin-expressing cells adhered more strongly to collagen and generated 3-fold higher traction forces compared with vinculin knockout cells. Moreover, vinculin-expressing cells were able to migrate into dense (5.8 mg/ml) three-dimensional collagen matrices that were impenetrable for vinculin knockout cells. These findings suggest that vinculin facilitates three-dimensional matrix invasion through up-regulation or enhanced transmission of traction forces that are needed to overcome the steric hindrance of ECMs.

Identifying New Potential Biomarkers in Adrenocortical Tumors Based on mRNA Expression Data Using Machine Learning.

Adrenocortical carcinoma (ACC) is a rare disease, associated with poor survival. Several "multiple-omics" studies characterizing ACC on a molecular level identified two different clusters correlating with patient survival (C1A and C1B). We here used the publicly available transcriptome data from the TCGA-ACC dataset (n = 79), applying machine learning (ML) methods to classify the ACC based on expression pattern in an unbiased manner. UMAP (uniform manifold approximation and projection)-based clustering resulted in two distinct groups, ACC-UMAP1 and ACC-UMAP2, that largely overlap with clusters C1B and C1A, respectively. However, subsequent use of random-forest-based learning revealed a set of new possible marker genes showing significant differential expression in the described clusters (e.g., SOAT1, EIF2A1). For validation purposes, we used a secondary dataset based on a previous study from our group, consisting of 4 normal adrenal glands and 52 benign and 7 malignant tumor samples. The results largely confirmed those obtained for the TCGA-ACC cohort. In addition, the ENSAT dataset showed a correlation between benign adrenocortical tumors and the good prognosis ACC cluster ACC-UMAP1/C1B. In conclusion, the use of ML approaches re-identified and redefined known prognostic ACC subgroups. On the other hand, the subsequent use of random-forest-based learning identified new possible prognostic marker genes for ACC.

Structural Analysis of the Caenorhabditis elegans Dauer Larval Anterior Sensilla by Focused Ion Beam-Scanning Electron Microscopy.

At the end of the first larval stage, the nematode Caenorhabditis elegans developing in harsh environmental conditions is able to choose an alternative developmental path called the dauer diapause. Dauer larvae exhibit different physiology and behaviors from non-dauer larvae. Using focused ion beam-scanning electron microscopy (FIB-SEM), we volumetrically reconstructed the anterior sensory apparatus of C. elegans dauer larvae with unprecedented precision. We provide a detailed description of some neurons, focusing on structural details that were unknown or unresolved by previously published studies. They include the following: (1) dauer-specific branches of the IL2 sensory neurons project into the periphery of anterior sensilla and motor or putative sensory neurons at the sub-lateral cords; (2) ciliated endings of URX sensory neurons are supported by both ILso and AMso socket cells near the amphid openings; (3) variability in amphid sensory dendrites among dauers; and (4) somatic RIP interneurons maintain their projection into the pharyngeal nervous system. Our results support the notion that dauer larvae structurally expand their sensory system to facilitate searching for more favorable environments.