Spatialproteomics: an interoperable toolbox for analyzing highly multiplexed fluorescence image data
TL;DR
This work demonstrates an end-to-end analysis from raw images to statistical characterization of how cell type composition and spatial distribution vary across indolent and aggressive lymphomas and presents a Python-based toolbox that supports end-to-end analysis of highly multiplexed imaging data.
Abstract
Highly multiplexed immunofluorescence imaging visualizes and quantifies protein levels at single-cell resolution in intact tissues at low cost and high scalability. Analysis of these data involves multiple steps with many method and parameter choices that must be adapted to the data and analytical objectives. There is an unmet need for a toolbox that offers flexible end-to-end coverage of the workflow. Here we present ‘spatialproteomics’, a Python package that addresses these challenges. Spatialproteomics enables the processing and analysis of large imaging data, including steps such as segmentation, image processing and cell-type classification, while synchronizing shared coordinates across data modalities. We demonstrate spatialproteomics on images of reactive lymph nodes and B cell non-Hodgkin lymphomas from 132 patients. We showcase an end-to-end analysis from raw images to statistical characterization of how cell type composition and spatial distribution vary across indolent and aggressive lymphomas. Furthermore, we show how spatialproteomics can process Gigapixel whole-slide images. Spatialproteomics is a Python-based toolbox that supports end-to-end analysis of highly multiplexed imaging data.