Yuan Li

H2O predicts spatial transcriptomics and proteomics from H&E images using a contrastive learning framework, outperforming existing models and recovering biologically relevant cell-cell communication.

The paper introduces SCPro, a multimodal spatial proteomics method, to profile pancreatic tumor heterogeneity by integrating imaging, flow cytometry, and protein aggregation capture, revealing spatial proteome variations and a novel regulatory T cell subtype.