A Python toolbox for analyzing multiplexed imaging data, featuring segmentation, pixel/cell clustering, and spatial analysis.
ark-analysis is a Python-based toolbox for analyzing multiplexed imaging data, providing an integrated pipeline for cell segmentation, pixel and cell clustering, and spatial analysis. It solves the problem of extracting quantitative, single-cell and spatial information from high-plex tissue images, such as those generated by MIBI or similar technologies. The pipeline guides users from raw image data to biological insights through a series of Jupyter notebooks.
Researchers and bioinformaticians working with multiplexed tissue imaging data, particularly in immunology, oncology, and spatial biology. It is suited for those needing to analyze high-plex images to identify cell phenotypes and their spatial relationships.
Developers choose ark-analysis for its complete, end-to-end workflow specifically designed for multiplexed imaging, its integration with state-of-the-art tools like Mesmer and Pixie, and its practical, notebook-based approach that combines automation with opportunities for manual refinement and visualization.
Integrated pipeline for multiplexed image analysis
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Provides a complete pipeline from segmentation with Mesmer to spatial analysis like pairwise enrichment, all within cohesive Jupyter notebooks as shown in the pipeline flowchart.
Incorporates deep learning segmentation (Mesmer) and advanced clustering (Pixie), both backed by peer-reviewed publications cited in the README, ensuring methodologically robust analysis.
Seamlessly integrates with Mantis Viewer for overlaying analysis results on raw images, and includes a ready-to-use example dataset with 11 FOVs for testing and validation.
Notebook-based structure allows researchers to follow a standardized workflow while permitting custom adjustments, enhancing reproducibility in biological studies as emphasized in the philosophy section.
Requires Conda environment creation, specific installation steps for Windows with additional guidance, and manual handling of Jupyter notebooks, which can be a barrier for users unfamiliar with command-line tools.
Relies heavily on Jupyter notebooks for execution, making it less suited for automated pipelines or integration into production environments without significant scripting, as updates require manual notebook duplication and environment refreshes.
Optimized for MIBI data and Linux/MacOS systems, with the README noting that Windows requires extra steps and other imaging platforms might need adaptation, limiting out-of-the-box usability.