Pseudo-spectral angle mapping for automated pixel-level analysis of highly multiplexed tissue image data

Abstract

The rapid development of highly multiplexed microscopy systems has enabled the study of cells embedded within their native tissue, which is providing exciting insights into the spatial features of human disease [1]. However, computational methods for analyzing these high-content images are still emerging, and there is a need for more robust and generalizable tools for evaluating the cellular constituents and underlying stroma captured by high-plex imaging [2]. To address this need, we have adapted spectral angle mapping – an algorithm used widely in hyperspectral image analysis – to compress the channel dimension of high-plex immunofluorescence images. As many high-plex immunofluorescence imaging experiments probe unique sets of protein markers, existing cell and pixel classification models do not typically generalize well. Pseudospectral angle mapping (pSAM) uses reference pseudospectra – or pixel vectors – to assign each pixel in an image a similarity score to several cell class reference vectors, which are defined by each unique staining panel. Here, we demonstrate that the class maps provided by pSAM can directly provide insight into the prevalence of each class defined by reference pseudospectra. In a dataset of high-plex images of colon biopsies from patients with gut autoimmune conditions, sixteen pSAM class representation maps were combined with instance segmentation of cells to provide cell class predictions. Finally, pSAM detected a diverse set of structure and immune cells when applied to a novel dataset of kidney biopsies imaged with a 43-marker panel. In summary, pSAM provides a powerful and readily generalizable method for evaluating high-plex immunofluorescence image data.Significance StatementUnderstanding the cellular constituents captured by highly multiplexed tissue imaging is a major limitation affecting the usability of these novel imaging methods. Many imaging experiments have uniquely designed staining panels, reducing the generalizability of cell classification models to new datasets. We present pseudospectral angle mapping (pSAM), which can compress high-dimensional image data into class representations. We demonstrate that the class representations generated by pSAM can be used to interpret high-plex image data and guide cell classification. Importantly, we also demonstrate that pSAM can generalize to new image datasets—collected with a different staining panel in samples from different tissues—without manual image annotation, subjective intensity gating, or re-training an algorithm.

Pixel-level analysis of protein markers in highly multiplexed immunofluorescence images of kidney biopsies

“…Given any high-plex staining panel, reference pseudospectra can be developed for evaluating image data pixel-by-pixel. We have previously shown in another dataset that these pixel-level class representations can also aid in cell phenotyping [21]. pSAM relies on libraries of pseudospectra for comparison rather than manual image annotations, alleviating the need for cell class ground truth, which can be very noisy in high-plex image data.…”

“…We have previously developed methods for compressing the channel dimension of highly multiplexed images, termed pseudo-spectral angle mapping (pSAM) [21]. Briefly, reference pseudo-spectra for immune cell populations of interest were developed with collaborators in rheumatology using 30 of the 43 immunofluorescence markers in the staining panel.…”

Pseudo-spectral angle mapping to improve immune cell classification in highly multiplexed fluorescence microscopy images

“…PSC is a rare autoimmune disease of the liver and is often associated with IBD [14,15]. Methods for data collection and preprocessing were previously described [13]. For clarity, these methods are briefly described below.…”

“…The resulting class maps were combined with whole cell and nucleus instance segmentations to classify detected cells as described in Durkee et. al [13].…”

“…The images produced by these high-plex imaging modalities are not unlike hyperspectral images. Previously, we demonstrated a method inspired by spectral angle mapping -a common method for analyzing hyperspectral images -can help to compress multi-channel immunofluorescence image into single-channel class representations [13]. This method, coined pseudo-spectral angle mapping (pSAM), produces likelihood maps for pre-defined cell classes of interest probed in high-plex IF experiments.…”