Multiple-choice microscopy pipeline
An end-to-end processing pipeline to transform large, multi-channel whole slide images into single-cell data. This site is a consolidated source of information on MCMICRO, documentation, roadmap, community and test data.
Images to Insights 💡
Highly multiplexed tissue imaging provides deep insight into the composition, organization and states of normal and diseased tissues. When converted into single cell data, tissue images are a natural complement to scRNA-Seq and similar profiling methods with the added advantage of spatial context. MCMICRO converts raw images into single cell data using state of the art algorithms for illumination correction, stitching, quality control, segmentation, and cell type calling.
MCMICRO Pipeline 🎛️
MCMICRO is open source, community supported software that uses Docker and workflow software to create pipelines for analyzing microscopy-based images of tissues, with an emphasis on highly multiplexed methods and single-cell data. Data is processed sequentially using algorithms (modules) developed in different research groups.
Flexible tools for a new field 🤝
High-plex tissue imaging is a new field involving a wide range of imaging technologies and the best image analysis approach is not always clear. MCMICRO therefore implements a “multiple choice” approach in which users can select among different modules for critical processing steps
Test Data 📥
MCMICRO comes with a growing library of imaging data (EMIT data) for testing your test run or for developing new algorithms. There is a lot of unexplored biology in the test data as well!
Technology Agnostic 🎩
MCMICRO works with any image that meets the BioFormats standard, most commonly OME-TIFF. These images can be acquired using a wide range of technologies- CODEX, CyCIF, mIHC, mxIF, IMC or MIBI.
Best Practices 🤓
Although it is nice to have many modules to try out, often you just want something that works. The MCMICRO team is collaborating with the NCI to run hackathons and challenges to identify the best modules and pipelines for specific types of data.