An end-to-end workflow for multiplexed image processing and analysis

Author(s): Jonas Windhager, Vito Zanotelli, Daniel Schulz, Lasse Meyer, Michelle Daniel, Bernd Bodenmiller, Nils Eling

Affiliation(s): University of Zurich

Social media: https://twitter.com/NilsEling

Highly multiplexed imaging allows the detection of dozens of biomolecules in single cells across tissue sections. Extracting biologically relevant information such as the spatial distribution of cell phenotypes from multiplexed tissue imaging data involves a number of computational tasks, including image segmentation, feature extraction, and spatially resolved single-cell analysis. To facilitate user-friendly and scalable analysis of data generated by multiplexed imaging technologies we developed a number of computational tools that integrate to form an end-to-end workflow for bioimage and spatially resolved single-cell analysis. For data quality assessment, we developed napari-imc, a plugin for the napari viewer, for interactively inspecting raw imaging data and the cytomapper R/Bioconductor package for image visualization in R. The steinbock framework supports image pre-processing, segmentation, feature extraction, and data export in a reproducible fashion. Single-cell, spatial data analysis such as community analysis, cellular neighborhood detection and cell-cell interaction testing are facilitated by the imcRtools R/Bioconductor package. Together, the developed tools support all relevant analysis steps for data generated by highly multiplexed imaging technologies including IMC, MIBI, CODEX, CyCIF and IBEX. We further developed these tools to extract biologically and clinically relevant features from images in a user-friendly and reproducible fashion.