Software Enables Researchers to Decode Lymphoma’s Cellular Architecture

Cancer outcomes are shaped not just by malignant cells, but by the spatial ecosystem around them—immune neighborhoods, vasculature, and stromal structure. In B-cell non-Hodgkin lymphoma, the arrangement of these components can tip disease toward growth, immune escape, or therapeutic response. Yet translating this three-dimensional context into quantitative biology has been a persistent technical bottleneck.

A collaborative team from the European Molecular Biology Laboratory (EMBL), Heinrich Heine University Düsseldorf (HHU), and University Hospital Düsseldorf (UKD) has introduced spatialproteomics, an open-source software package aimed at simplifying analysis of highly multiplexed tissue imaging. The work appears in Nature Methods.

Multiplexed fluorescence imaging can label dozens of molecular markers while preserving where each cell sits in a tissue section. With this capability, researchers can survey thousands to millions of cells and map the tumor microenvironment with far greater depth than conventional pathology. The payoff is an unprecedented view of protein distributions across cell types and microanatomical structures.

The challenge begins after imaging. Before spatial relationships can be assessed, investigators must segment cells, identify and annotate them, and quantify protein signals across massive datasets. Historically, these tasks often required stitching together multiple tools and custom pipelines, reducing reproducibility and raising barriers for new users.

Spatialproteomics addresses these steps within a single interoperable workflow, guiding users from raw microscopy inputs through segmentation, annotation, and spatial analysis. By standardizing the computational route from image to metric, the software is designed to make results more consistent across studies and more scalable to large cohorts.

To validate the framework, the team analyzed patient samples from B-cell non-Hodgkin lymphoma. The software enabled systematic characterization of how proteins distribute across tumor and immune compartments, revealing distinct spatial patterns tied to disease biology. These computational signatures helped differentiate indolent, slow-growing lymphomas from more aggressive forms.

The study also underscores a broader shift in biomedical imaging: as technologies generate richer spatial data, the limiting factor moves toward accessible, robust analysis software. By lowering technical hurdles and emphasizing reproducible workflows, spatialproteomics supports wider adoption of spatially resolved proteomic insights.

Because lymphoma is only one application, the authors anticipate benefits across many areas of biomedical research, including other cancers and inflammatory disorders where tissue organization is central to outcome.