CellNiche represents cellular microenvironments in atlas-scale spatial omics data with contrastive learning

Abstract

Abstract Deciphering cellular microenvironments at atlas scale remains challenging because molecular identity, spatial context, and platform heterogeneity are tightly coupled. Here we present CellNiche, a scalable contrastive-learning framework that identifies and characterizes cellular microenvironments from spatial omics data using cell-centric spatial-proximity subgraphs. CellNiche combines spatial co-localization and molecular co-expression cues to learn microenvironment-aware embeddings. Across spatial omics datasets from multiple platforms (>10 million cells in total), scaling experiments show improved representations with more training data and competitive clustering and embedding-quality performance with efficient computation. In a multi-sample human non-small-cell lung cancer (NSCLC) cohort, CellNiche identifies conserved and sample-specific tumor and immune microenvironments and captures localized spatial transitions. In four independent mouse brain atlases, CellNiche integrates 293 slices into a unified virtual brain map for cross-atlas annotation transfer and spatial refinement.