Stepwise multi-scale reconstruction of cell spatial organization from single-cell RNA sequencing data with Cell2space.

Cell2space accurately reconstructs multi-scale tissue architecture from scRNA-seq and spatial transcriptomics data, achieving superior domain and neighborhood inference.

• Why it matters: Understanding tissue organization is crucial for insights into function and disease, but dissociation during scRNA-seq loses spatial context, creating a significant gap in data integration.
• What they did: The framework employs deep learning to learn a universal spatial affinity function and uses hierarchical inference with domain priors, integrating scRNA-seq and ST references across multiple scales.
• The result: Cell2space successfully identifies tissue layers, gene expression gradients, and microenvironments in mouse cortex and human skin, enabling detailed tissue architecture reconstruction without prior annotations.