Search GIDE databases

The cell cycle governs proliferation of all eukaryotic cells. Profiling cell cycle dynamics is therefore central to basic and biomedical research. However, current approaches to cell cycle profiling involve complex interventions that may confound experimental interpretation. We developed CellCycleNet, a machine learning (ML) workflow, to simplify cell cycle staging from fluorescent microscopy data with minimal experimenter intervention and cost. CellCycleNet accurately predicts cell cycle phase using only a fluorescent nuclear stain (DAPI) in fixed interphase cells. Using the Fucci2a cell cycle reporter system as ground truth, we collected two benchmarking image datasets and trained 2D and 3D ML modelsÑof support vector machine (SVM) and deep neural network architectureÑto classify nuclei in the G1 or S/G2 phases. Our results show that 3D CellCycleNet outperforms SVM models on each dataset. When trained on two image datasets simultaneously, CellCycleNet achieves the highest classification accuracy (AUROC of 0.94-0.95). Overall, we found that using 3D features, rather than 2D features alone, significantly improves classification performance for all model architectures. We released our image data, models, and software as a community resource.

Imaging Methods: confocal microscopy fluorescence microscopy

Organisms: Mus musculus

Phospholipidosis (PLD) is a cellular adverse effect that is, among others, caused by numerous cationic amphiphilic drugs. Interest is raised within pharma discovery to predict this phenomenon, as it can impact the outcome of phenotypic cellular screens and significantly delay drug development processes. The development of accurate and validated machine learning models for predicting drug-induced PLD across different cell lines and research centers could provide a valuable early application tool for the pharmaceutical industry, potentially accelerating drug discovery and reducing the risk of late-stage failures. We report here the assembly, curation, testing and modeling of one of the largest datasets of repurposed drugs (5000+) tested for PLD induction on different cell lines. A machine-learning classification method was developed and validated to predict whether molecules are prone to induce PLD effects when applied in cell-based screens.

Imaging Methods: spinning disk confocal microscopy

Organisms: Chlorocebus aethiops

Images of cultured cells, organoids, glia and neurons in brain slices and in the brains of mice.

Imaging Methods: recorded image

Organisms: Mus musculus Homo sapiens

Promyelocytic leukemia (PML) proteins are known to form phase-separated nuclear punctate structures called PML-nuclear bodies (PML-NBs). The integrity disruption of PML-NBs is linked with the pathogenesis of acute promyelocytic leukemia (APL), and trivalent arsenic (As3+) has been used for the clinical treatment of APL to restore normal PML-NBs. As3+ is considered to bind to cysteine residues and enhances modification of PML with small-ubiquitin-like protein (SUMO). We exposed U-2OS and CHO-K1 cells stably overexpressing PML-VI to As3+ and found that the solubility of PML decreased and SUMOylation of PML increased after 2 h-exposure to 3 μM As3+. Contrary to As3+-induced remarkable biochemical changes including the solubility change and SUMOylation of PML, microscopic observation of PML-NBs was not changed clearly after a short-term exposure to As3+. The number of PML-NBs decreased and extranuclear PML bodies (EnPBs), which are remniscences of PML-NBs after nuclear membrane breakdown at mitosis, increased after exposure to As3+ for 24 - 72 h. The amount of SUMOylated PML decreased after prolonged exposure to As3+ while the solubility of PML was kept low, suggesting that As3+ stabilized EnPB without SUMOylation. The effects of As3+ on EnPBs were clearly observed at as low as 0.3 μM As3+ which corresponds to inorganic arsenic level in drinking water worldwide. Keywords: arsenic, small ubiquitin-like modifier (SUMO), solubility, phase separation, promyelocytic leukemia-nuclear body (PML-NB), extranuclear PML body (EnPB)

Imaging Methods: recorded image

Organisms: Homo sapiens Cricetulus griseus

The dataset provides fluorescence images of myocytes in subepicardial myocardium of Langendorff-perfused rat hearts, which were rapidly-frozen (RF) during different contractile state including systole, diastole, and ventricular fibrillation (VF). The RF hearts were subsequently fixed with acetone and paraformaldehyde (PFA) by freeze-substitution procedure, then α-actinin in myocytes of them was immuno-stained with fluorescence labels. As control, hearts rapidly-frozen under mechanical relaxation induced by 2,3-butanedione monoxime (BDM), hearts fixed with PFA solely, hearts rapidly-frozen after PFA-fixation were also prepared.

Imaging Methods: electron microscopy recorded image

Organisms: Rattus norvegicus

Images of neurons in the cerebral cortex in mice, which are labeled with fluorescent proteins.

Imaging Methods: recorded image

Organisms: Mus musculus

2D and 3D instance segmentation of nuclei from array tomography (AT) and FIB-SEM datasets, respectively. In this deposition, 10 AT and 2 FIB-SEM datasets and paired annotations are included. Note: in several datasets, partial annotations of nuclei at image edges have been deleted. These are indicated in the annotation metadata.

Imaging Methods: scanning electron microscopy (SEM) focussed ion beam scanning electron microscopy (FIB-SEM)

Organisms: Homo sapiens Mus musculus

Calcium transients drive cells to discharge prostaglandin E2 (PGE2). The authors visualized PGE2-induced protein kinase A (PKA) activation and quantitated PGE2 secreted from a single cell by combining fluorescence microscopy and a simulation model. For this purpose, they first prepared PGE2-producer cells that express either an optogenetic or a chemogenetic calcium channel stimulator: OptoSTIM1 (CRY2clust) or Gq-DREADD, respectively. Second, they prepared reporter cells expressing the Gs-coupled PGE2 reporter EP2 and the PKA biosensor Booster-PKA, which is based on the principle of Forster resonance FRET. Upon the stimulation-induced triggering of calcium transients, a single producer cell discharges PGE2 to stimulate PKA in the surrounding reporter cells. Due to the flow of the medium, the PKA-activated area exhibited a comet-like smear when HeLa cells were used. In contrast, radial PKA activation was observed when confluent MDCK cells were used, indicating that PGE2 diffusion was restricted to the basolateral space. By fitting the radius of the PKA-activated area to a simulation model based on simple diffusion, the authors estimated that a single HeLa cell secretes 0.25 fmol PGE2 upon a single calcium transient to activate PKA in more than 1000 neighboring cells. This model also predicts that the PGE2 discharge rate is comparable to the diffusion rate. Thus, their method quantitatively envisions that a single calcium transient affects more than 1000 neighboring cells via PGE2.

Imaging Methods: recorded image fluorescence microscopy time lapse microscopy FRET

Organisms: Homo sapiens Canis lupus familiaris

In the pistil of flowering plants, each ovule usually associates with a single pollen tube for fertilization. This one-to-one pollen tube guidance, which contributes to polyspermy blocking and efficient seed production, is largely different from animal chemotaxis of many sperms to one egg. However, the functional mechanisms underlying the directional cues and polytubey blocks in the depths of the pistil remain unknown. Here, the authors develop a two-photon live imaging method to directly observe pollen tube guidance in the pistil of Arabidopsis thaliana, clarifying signaling and cellular behaviors in the one-to-one guidance. Ovules are suggested to emit multiple signals for pollen tubes, including an integument- dependent directional signal that reaches the inner surface of the septum and adhesion signals for emerged pollen tubes on the septum. Not only FERONIA in the septum but ovular gametophytic FERONIA and LORELEI, as well as FERONIA- and LORELEI- independent repulsion signal, are involved in polytubey blocks on the ovular funiculus. However, these funicular blocks are not strictly maintained in the first 45 min, explaining previous reports of polyspermy in flowering plants.

Imaging Methods: recorded image two-photon laser scanning microscopy time lapse microscopy

Organisms: Arabidopsis thaliana

Plants develop along apical-basal and radial axes. In Arabidopsis thaliana, the radial axis becomes evident when the cells of the 8-cell proembryo divide periclinally, forming inner and outer cell layers. Although changes in cell polarity or morphology likely precede this oriented cell division, the initial events and the factors regulating radial axis formation remain elusive. Here, we report that three transcription factors belonging to the class IV homeodomain-leucine zipper (HD-ZIP IV) family redundantly regulate radial pattern formation:HOMEODOMAIN GLABROUS11 (HDG11), HDG12, and PROTODERMAL FACTOR2 (PDF2). The hdg11 hdg12 pdf2 triple mutant failed to undergo periclinal division at the 8-cell stage and cell differentiation along the radial axis. Live-cell imaging revealed that the mutant defect is already evident in the behavior of the embryo’s initial cell (apical cell), which is generated by zygote division. In the wild type, the apical cell grows longitudinally and then radially, and its nucleus remains at the bottom of the cell, where the vertical cell plate emerges. By contrast, the mutant apical cell elongates longitudinally, and its nucleus releases from its basal position, resulting in a transverse division. Computer simulations based on the live-cell imaging data confirmed the importance of the geometric rule (the minimal plane principle and nucleus-passing principle) in determining the cell division plane. We propose that HDG11, HDG12, and PDF2 promote apical cell polarization, i.e., radial cell growth and basal nuclear retention, and set proper radial axis formation during embryogenesis.

Imaging Methods: recorded image spinning disk confocal microscopy spinning disk confocal microscopy

Organisms: NCBITaxon:3702

This dataset contains high-content, live-cell, time-lapse imaging of HeLa cells undergoing apoptosis across a 10-point dose response to the apoptosis inducer staurosporine. Using the ChromaLIVE™ Live Cell Painting assay, cells were imaged every 30 minutes over a six-hour period on the Yokogawa CellVoyager CQ1 spinning-disk confocal platform, capturing multiplexed fluorescence signals (Hoechst, ChromaLIVE 488_yellow, ChromaLIVE 488_red, and ChromaLIVE 561) across three z-planes at four fields of view per well. After live imaging, cells were fixed and stained with AnnexinV to provide a single-cell ground-truth readout of apoptosis. The dataset includes raw and illumination-corrected images, cell segmentation masks, single-cell tracking outputs, and extracted morphological features generated using CellProfiler and scDINO deep-learning embeddings. In total, the dataset comprises over 188,000 single-cell observations and 20,677 cell tracks across time, with 2,336 curated morphology features per cell. These measurements quantify the dynamic morphological signatures preceding apoptosis and capture dose- and time-dependent phenotypic trajectories. This resource enables the study of temporal morphology changes, benchmarking of image-based profiling pipelines, development of machine-learning models for early apoptosis prediction, and exploration of dynamic cell-state transitions. It supports future work such as profiling additional forms of regulated cell death and constructing temporal morphology atlases.

Organisms: Homo sapiens

Plasma membrane calcium ATPases (PMCAs) are believed to function exclusively at the plasma membrane where they expel calcium from the cytosol. We have unexpectedly identified a splice variant-dependent localisation of the PMCA isoform PMCA2 to the lysosome, where it forms an evolutionarily conserved complex with NPC1, the lysosomal membrane protein defective in the rare lysosomal storage disease Niemann-Pick disease type C (NPC). This interaction is required for lysosomal Ca2+ homeostasis and implicates PMCA2 as a mediator of Ca2+ uptake into lysosomes. Disruption of the NPC1-PMCA2 complex contributes to the pathophysiology of both Niemann-Pick disease type C and Parkinson’s disease, revealing an unrecognised intracellular function for PMCA2 and a shared mechanism linking lysosomal Ca2+ and lipid regulation in neurodegeneration.

Imaging Methods: confocal microscopy

Organisms: Rattus norvegicus