Software

MCMICRO is an open-source processing pipeline for multiplexed whole-tissue images and tissue microarrays. It consists of customizable modules that sequentially perform image processing and quantification steps, including stitching, registration, cell segmentation, single-cell quantification, and visualization. Each module is containerized with Docker, making it possible to deploy MCMICRO across various computing environments, including local machines, job-scheduling clusters, and cloud environments like AWS. MCMICRO is undergoing active development, and modules are regularly added and improved as part of the growing MCMICRO community.

Minerva is a suite of lightweight software tools that enables interactive viewing and fast sharing of large image data. With Minerva Author, users can import an image and generate a multi-waypoint, annotated Minerva Story that walks viewers through their data. Minerva Stories are hosted on the web and viewed through a web browser (no download required!), making them ideal for sharing large-scale image data as part of tissue atlases.

CyLinter is quality control software for multiplexed microscopy images that identifies and removes single cell data that has been corrupted by artifacts. Artifacts can result from optical aberrations (e.g., from contaminating lint or hair on the sample) or image-processing errors (e.g., errors in single cell segmentation) and can confound downstream analyses. CyLinter guides users through the process of removing noisy data in an interactive, visual environment and improves the quality of the resulting single cell data.

SCIMAP is a scalable toolkit for analyzing spatial molecular data. SCIMAP is compatible with spatial datasets mapped to X-Y coordinates and can be integrated with many single-cell analysis toolkits. SCIMAP supports preprocessing, phenotyping, visualization, clustering, spatial analysis, and differential spatial testing, and can efficiently analyze large datasets containing millions of cells.

ASHLAR (Alignment by Simultaneous Harmonization of Layer/Adjacency Registration) is an open-source Python tool that combines multi-tile microscopy images into high-dimensional mosaic images. For multi-cycle imaging methods (like CyCIF), ASHLAR also aligns images from different cycles with a high level of accuracy. ASHLAR can be used with virtually any unstitched microscope image file and multiplexed imaging method.

Identifying single cells within tissue images, a process known as segmentation, is critical for single-cell analysis of imaging data. UnMICST (Universal Models for Identifying Cells and Segmenting Tissue) is a semantic segmentation method for tissue images. UnMICST uses a convolutional neural network to generate probability maps that classify whether a given pixel belongs to the foreground or background class. UnMICST was trained using images of real human tissue, including real augmentations (e.g., out-of-focus images) and nuclear envelope stains, to improve segmentation accuracy across a wide range of tissue types. The latest version of UnMICST is part of the MCMICRO pipeline.

Scope2Screen is a scalable software system for focus+context exploration and annotation of whole-slide, high-plex, tissue images. Our approach scales to analyzing 100GB images of 10^9 or more pixels per channel, containing millions of individual cells. A multidisciplinary team of visualization experts, microscopists, and pathologists identified key image exploration and annotation tasks involving finding, magnifying, quantifying, and organizing regions of interest (ROIs) in an intuitive and cohesive manner.