The dual-enhanced graph learning framework DePass allows paired data integration in single-cell and spatial multiomics.

DePass achieves superior multi-omics data integration across six modalities and 13 platforms, revealing detailed tissue heterogeneity at near single-cell resolution.

• Why it matters: Effective integration of noisy, multimodal data is essential for understanding cellular and tissue complexity, yet current methods are limited to single data types and often overlook noise challenges.
• What they did: DePass employs a coupled enhancement-integration graph learning framework that iteratively denoises data and refines embeddings, tested systematically across diverse biological contexts.
• The result: The framework enables accurate, scalable integration, uncovering immune niche substructures and tumor heterogeneity, advancing multi-omics analysis in single-cell and spatial studies.