SpaceBlocks: an expert-supervised workflow for cohort-level spatial transcriptomics analysis

SpaceBlocks, a Snakemake workflow for single-cell resolution spatial transcriptomics, reanalyzed 37 Visium HD, Xenium 5K, and MERSCOPE samples with scalable processing across platforms.

• Why it matters: Spatial transcriptomics data vary by technology, and reproducible cohort-level pipelines often lack expert-supervised decisions that preserve computational and biological provenance.
• What they did: SpaceBlocks homogenizes heterogeneous platform outputs into a validated common format, then uses per-sample quality control and cell type annotation diagnostic visualizations to guide expert supervision before automated downstream steps.
• The result: In the 37-sample reanalysis, it recovered previously missed cell types in Visium HD, found heterogeneous marker gene expression in MERSCOPE, and enabled domain-linked differential expression for multiple sclerosis and healthy–malignant differences in Visium HD.

AI summary of the abstract · may contain errors · not medical advice · read the paper.