Esselman, A. B. | Search bioRxiv

bioRxiv2025

Unmixing of Imaging Mass Spectrometry Measurements Using Microscopy-Informed Constraints

Imaging mass spectrometry (IMS) provides spatially resolved molecular information of organic tissue but can be limited by pixel signals mixing contributions from adjacent biological structures, e.g. of single cells and multicellular functional tissue units (FTUs). This paper proposes computational methods to predict mass spectral profiles of biological structures on the basis of IMS data by "unmixing" pixel-level signals, leveraging microscopy-based boundary information of these structures. By modeling each biological structure as having a unique mass spectrum, we formulate a linear mixing model and solve the corresponding inverse problem that unmixes blended signals. In particular, we cover both overdetermined and underdetermined linear system scenarios and compare ordinary least squares, nonnegative least squares, and singular value thresholding to a custom algorithm, coined Tissue-informed Unmixing of Labeled regions by Inverse Problem (TULIP), specifically tailored to IMS data. Validation on a synthetic in-situ single cell dataset and demonstration on a largescale kidney FTU dataset illustrate the potential of these methods for enhanced in-situ tissue structure analysis, e.g. in cellular and tissue studies.

bioinformatics↗

bioRxiv2024

Revealing In Situ Molecular Profiles of Glomerular Cell Types and Substructures with Integrated Imaging Mass Spectrometry and Multiplexed Immunofluorescence Microscopy

Glomeruli filter blood through the coordination of podocytes, mesangial cells, fenestrated endothelial cells, and the glomerular basement membrane. Cellular changes, such as podocyte loss, are associated with pathologies like diabetic kidney disease (DKD). However, little is known regarding the in situ molecular profiles of specific cell types and how these profiles change with disease. Matrix-assisted laser desorption/ionization imaging mass spectrometry (MALDI IMS) is well-suited for untargeted tissue mapping of a wide range of molecular classes. Additional imaging modalities can be integrated with MALDI IMS to associate these biomolecular distributions to specific cell types. Herein, we demonstrate an integrated workflow combining MALDI IMS and multiplexed immunofluorescence (MxIF) microscopy. High spatial resolution MALDI IMS (5 {micro}m pixel size) was used to determine lipid distributions within human glomeruli, revealing intra-glomerular lipid heterogeneity. Mass spectrometric data were linked to specific glomerular cell types through new methods that enable MxIF microscopy to be performed on the same tissue section following MALDI IMS without sacrificing signal quality from either modality. A combination of machine-learning approaches was assembled, enabling cell-type segmentation and identification based on MxIF data followed by the mining of cell type or cluster-associated MALDI IMS signatures using classification models and interpretable machine learning. This allowed the automated discovery of spatially specific biomarker candidates for glomerular substructures and cell types. Overall, the work presented here establishes a toolbox for probing molecular signatures of glomerular cell types and substructures within tissue microenvironments and provides a framework that applies to other kidney tissue features and organ systems.

biochemistry↗