Visvikis, T. | bioRxiv Science

bioRxiv2026

Thyra: Bridging Mass Spectrometry Imaging and SpatialData for Unified Multi-Modal Analysis

Mass Spectrometry Imaging (MSI) is a powerful technique for mapping molecular distributions, and its integration with other imaging modalities is crucial for comprehensive understanding of molecular systems. Fragmented data formats and the limitations of existing standards like imzML, challenge spatial biology centric multi-modal data analysis and adherence to FAIR data principles. This paper introduces Thyra, a modern Python library designed to convert MSI data into the SpatialData framework, a unified and extensible multi-platform file format that crucially integrates MSI into the broader spatial omics ecosystem. Thyras modular architecture, intelligent mass axis resampling, and sparse matrix backend address performance bottlenecks and facilitate seamless integration with tools for advanced spatial statistics and visualization. By adopting SpatialData, Thyra not only improves data interoperability, accessibility, and reusability, but also unlocks new avenues for multi-modal research, empowering scientists to integrate rich chemical information into diverse biological workflows.

bioinformatics↗

bioRxiv2025

One Section, Two Worlds: Single-Cell Integration of MALDI-MSI and Spatial Transcriptomics on the Same Single Tissue Section

Understanding tissue complexity requires spatially resolved multiomic data at single-cell resolution. Here, we present a workflow that integrates high-resolution matrix-assisted laser desorption ionization mass spectrometry imaging (MALDI-MSI) with Xenium spatial transcriptomics (SPT) on a single tissue section. This one-section strategy ensures exact spatial correspondence between metabolic and transcriptomic features, avoiding the misalignment issues of serial sections, where even minor offsets can result in sampling different cells. We validate compatibility of MALDI-MSI with downstream SPT, preserving transcriptomic quality despite semi-destructive ionization. Using mouse brain and human glioblastoma tissues, we achieve pixel-perfect modality coregistration, enabling per-cell MALDI spectra extraction aligned with gene expression. Integrated clustering reveals enhanced cell-type resolution and identifies metabolic heterogeneity within transcriptionally defined populations. This enables a direct and precise correlation between what a cell is doing and its biochemical state, providing a more holistic and accurate picture of cellular function, heterogeneity, and interaction in health and disease. Our workflow provides a scalable path to multiomic atlases of disease and development, advancing both data integration and translational research.

molecular biology↗