Umeå University's logo

Accurate cell tracking in microscopy is essential for studying biological dynamics like proliferation and migration. Traditional fully supervised methods demand dense pixel-wise masks for every frame, making them impractical for large-scale use. Recent methods like SAT reduce annotation effort by using sparse point-based supervision, but still require multiple positive and negative points per cell, which remains labor-intensive. BoxTrack offers a lightweight and annotation-efficient alternative, requiring only a single bounding box per cell in the first frame. Without relying on any point-level annotations, it performs end-to-end instance segmentation and tracking over entire sequences. This simplification leads to a substantial reduction in annotation cost while improving performance over SAT. On the CTMC dataset, BoxTrack improves Multiple Object Tracking Accuracy (MOTA) by +15.96% over SAT. For the CTC dataset, it yields a +8.86% MOTA gain. Code is available at https://github.com/nabeelkhalid92/Box-it-Track-it.

Morphological profiling is a common approach to investigate the modes of action (MOAs) of compounds. Most methods rely on fixed-cell assays, which provide only a single snapshot at a predefined time point and overlook the dynamic nature of cellular responses. In contrast, live-cell imaging tracks responses over time, offering deeper insight into compound-specific effects and mechanisms; however, time-series analysis of image data remains challenging due to limited analytical tools. We present Live Cell Temporal Profiling (LCTP), a workflow for morphological profiling of label-free live-cell time series data that yields interpretable, biologically relevant results. We showcase LCTP in an MOA classification study using label-free data. The workflow integrates established deep-learning components, cell segmentation, live/dead classification, and single-cell feature extraction, with data-driven models to capture MOA-specific temporal phenotypes and produce time-resolved profiles that can be compared across compounds and cell lines. We assess MOA classification performance using double-blinded cross-validation simulating a real-world screening scenario. LCTP significantly improves MOA classification over single–time point analysis, consistently across both cell lines used in the study. Time-resolved phenotypic modelling reveals transient, sustained, and delayed responses, clarifying compound-specific temporal effects and mechanisms across MOAs. The presented workflow is modular: each step removes irrelevant information, enriching signal, and enabling straightforward updates as technologies evolve and as new technologies become available, while supporting reuse across studies broadly. We believe LCTP adds substantial value to high-throughput compound screening, showing that live-cell imaging combined with this workflow yields informative visualizations of temporal effects and improved MOA classification.

Multiclass data sets and large-scale studies are increasingly common in omics sciences, drug discovery, and clinical research due to advancements in analytical platforms. Efficiently handling these data sets and discerning subtle differences across multiple classes remains a significant challenge. In metabolomics, two-class orthogonal projection to latent structures discriminant analysis (OPLS-DA) models are widely used due to their strong discrimination capabilities and ability to provide interpretable information on class differences. However, these models face challenges in multiclass settings. A common solution is to transform the multiclass comparison into multiple two-class comparisons, which, while more effective than a global multiclass OPLS-DA model, unfortunately results in a manual, time-consuming model-building process with complicated interpretation. Here, we introduce an extension of OPLS-DA for data-driven multiclass classification: orthogonal partial least squares-hierarchical discriminant analysis (OPLS-HDA). OPLS-HDA integrates hierarchical cluster analysis (HCA) with the OPLS-DA framework to create a decision tree, addressing multiclass classification challenges and providing intuitive visualization of interclass relationships. To avoid overfitting and ensure reliable predictions, we use cross-validation during model building. Benchmark results show that OPLS-HDA performs competitively across diverse data sets compared to eight established methods. This method represents a significant advancement, offering a powerful tool to dissect complex multiclass data sets. With its versatility, interpretability, and ease of use, OPLS-HDA is an efficient approach to multiclass data analysis applicable across various fields.

Integrating cell segmentation with tracking is critical for achieving a detailed and dynamic understanding of cellular behavior. This integration facilitates the study and quantification of cell morphology, movement, and interactions, offering valuable insights into a wide range of biological processes and diseases. However, traditional methods rely on labor-intensive and costly annotations, such as full segmentation masks or bounding boxes for each cell. To address this limitation, we present SAT: Segment and Track Anything for Microscopy, a novel pipeline that leverages point annotations in the first frame to automate cell segmentation and tracking across all subsequent frames. By significantly reducing annotation time and effort, SAT enables efficient and scalable analysis, making it well-suited for large-scale studies. The pipeline was evaluated on two diverse datasets, achieving over 80% Multiple Object Tracking Accuracy (MOTA), demonstrating its robustness and effectiveness across various imaging modalities and cell types. These results highlight SAT’s potential to streamline biomedical research and enable deeper exploration of cellular behavior.

Single-cell RNA-seq methods can be used to delineate cell types and states at unprecedented resolution but do little to explain why certain genes are expressed. Single-cell ATAC-seq and multiome (ATAC + RNA) have emerged to give a complementary view of the cell state. It is however unclear what additional information can be extracted from ATAC-seq data besides transcription factor binding sites. Here, we show that ATAC-seq telomere-like reads counter-inituively cannot be used to infer telomere length, as they mostly originate from the subtelomere, but can be used as a biomarker for chromatin condensation. Using long-read sequencing, we further show that modern hyperactive Tn5 does not duplicate 9 bp of its target sequence, contrary to common belief. We provide a new tool, Telomemore, which can quantify nonaligning subtelomeric reads. By analyzing several public datasets and generating new multiome fibroblast and B-cell atlases, we show how this new readout can aid single-cell data interpretation. We show how drivers of condensation processes can be inferred, and how it complements common RNA-seq-based cell cycle inference, which fails for monocytes. Telomemore-based analysis of the condensation state is thus a valuable complement to the single-cell analysis toolbox.

Biopharmaceuticals are medical compounds derived from biological sources and are often manufactured by living cells, primarily Chinese hamster ovary (CHO) cells. CHO cells display variation among cell clones, leading to growth and productivity differences that influence the product's quantity and quality. The biological and environmental factors behind these differences are not fully understood. To identify metabolites with a consistent relationship to productivity or cell death over time, we analyzed the extracellular metabolome of 11 CHO clones with different growth and productivity characteristics over 14 days. However, in bioreactor processes, metabolic profiles and process variables are both strongly time-dependent, confounding the metabolite-process variable relationship. To address this, we customized an existing hierarchical approach for handling time dependency to highlight metabolites with a consistent correlation to a process variable over a selected timeframe. We benchmarked this new method against conventional orthogonal partial least squares (OPLS) models. Our hierarchical method highlighted several metabolites consistently related to productivity or cell death that the conventional method missed. These metabolites were biologically relevant; most were known already, but some that had not been reported in CHO literature before, such as 3-methoxytyrosine and succinyladenosine, had ties to cell death in studies with other cell types. The metabolites showed an inverse relationship with the response variables: those positively correlated with productivity were typically negatively correlated with the death rate, or vice versa. For both productivity and cell death, the citrate cycle and adjacent pathways (pyruvate, glyoxylate, pantothenate) were among the most important. In summary, we have proposed a new method to analyze time-dependent omics data in bioprocess production. This approach allowed us to identify metabolites tied to cell death and productivity that were not detected with traditional models.

Examining boiler failure causes is crucial for thermal power plant safety and profitability. However, traditional approaches are complex and expensive, lacking precise operational insights. Although data-driven approaches hold substantial potential in addressing these challenges, there is a gap in systematic approaches for investigating failure root causes with unlabeled data. Therefore, we proffered a novel framework rooted in data mining methodologies to probe the accountable operational variables for boiler failures. The primary objective was to furnish precise guidance for future operations to proactively prevent similar failures. The framework was centered on two data mining approaches, Principal Component Analysis (PCA) + K-means and Deep Embedded Clustering (DEC), with PCA + K-means serving as the baseline against which the performance of DEC was evaluated. To demonstrate the framework’s specifics, a case study was performed using datasets obtained from a waste-to-energy plant in Sweden. The results showed the following: (1) The clustering outcomes of DEC consistently surpass those of PCA + K-means across nearly every dimension. (2) The operational temperature variables T-BSH3rm, T-BSH2l, T-BSH3r, T-BSH1l, T-SbSH3, and T-BSH1r emerged as the most significant contributors to the failures. It is advisable to maintain the operational levels of T-BSH3rm, T-BSH2l, T-BSH3r, T-BSH1l, T-SbSH3, and T-BSH1r around 527 °C, 432 °C, 482 °C, 338 °C, 313 °C, and 343 °C respectively. Moreover, it is crucial to prevent these values from reaching or exceeding 594 °C, 471 °C, 537 °C, 355 °C, 340 °C, and 359 °C for prolonged durations. The findings offer the opportunity to improve future operational conditions, thereby extending the overall service life of the boiler. Consequently, operators can address faulty tubes during scheduled annual maintenance without encountering failures and disrupting production.

Cell Painting is an established community-based microscopy-assay platform that provides high-throughput, high-content data for biological readouts. In November 2022, the JUMP-Cell Painting Consortium released the largest publicly available Cell Painting dataset with CellProfiler features, comprising more than 2 billion cell images. This dataset is designed for predicting the activity and toxicity of 115k drug compounds, with the aim to make cell images as computable as genomes and transcriptomes. In this context, our paper introduces a scalable and computationally efficient data analytics workflow created to meet the needs of researchers. This data-driven workflow facilitates the comparison of drug treatment effects through significant and biologically relevant insights. The workflow consists of two parts: first, the Equivalence score (Eq. score), a straightforward yet sophisticated metric highlighting relevant deviations from negative controls based on cell image morphology; second, the scalability of the workflow, by utilizing the Eq. scores on a large scale to predict and classify the subtle morphological changes in cell image profiles. By doing so, we show classification improvements compared to using the raw CellProfiler features on the CPJUMP1-pilot dataset on three types of perturbations. We hope that our workflow's contributions will enhance drug screening efficiency and streamline the drug development process. As this process is resource-intensive, every incremental improvement is valuable. Through our collective efforts in advancing the understanding of high-throughput image-based data, we aim to reduce both the time and cost of developing new, life-saving treatments.

Microscopic imaging plays a pivotal role in various fields of science and medicine, offering invaluable insights into the intricate world of cellular biology. At the heart of this endeavor lies the need for accurate identification and characterization of individual cells within these images. Deep learning-based cell segmentation, which involves delineating cells from complex microscopic images, is pivotal for cell analysis. It serves as the foundation for extracting meaningful information about cell morphology, spatial organization, and interactions. However, traditional deep-learning models for cell segmentation require extensive and expensive annotation masks for each cell in the image, posing a significant challenge. To address this issue, this study introduces CellBoxify, a novel pipeline that streamlines cell instance segmentation. Unlike traditional methods, CellBoxify operates solely on bounding box annotations, making it approximately seven times faster than manual segmentation mask annotation for each cell. The proposed approach’s effectiveness is evident in its performance on the LIVECell dataset, a well-known resource for cell segmentation research. Achieving 83.40% of the fully supervised performance on this dataset demonstrates the efficacy of the proposed method.

Cellular imaging plays a pivotal role in understanding various biological processes and diseases, making accurate cell segmentation indispensable for many biomedical applications. However, traditional methods for cell segmentation often rely on manual annotation, which is labor-intensive and time-consuming. Deep learning-based approaches for cell segmentation have shown promising results, but they require a vast amount of annotated data for training. In this context, this study presents CellGenie, an end-to-end pipeline designed to address the challenge of data scarcity in deep learning-based cell segmentation. This research proposes an innovative approach for automatic synthetic data generation tailored for microscopic image analysis. Leveraging the rich information provided by the LIVECell dataset, CellGenie generates synthetic microscopic images along with their corresponding segmentation masks for individual cells. By seamlessly integrating this synthetic data into the training process, this study enhances the performance of cell segmentation models beyond the limitations of existing annotated dataset. Furthermore, extensive experimentations are conducted to evaluate the efficacy of the generated data across various experimental scenarios. The results demonstrate the substantial impact of synthetic data generation in improving the robustness and generalization of cell segmentation models.