scComm: a contrastive learning framework for deciphering cell-cell communications at single-cell resolution (mIF)

We performed hierarchical clustering to the tumor cells using the interaction scores calculated by scComm and detected three primary tumor cell groups: tumor type 1(TT1) with heightened interactions with fibroblasts, tumor cells, and endothelial cells; another group (TT2) with increased interactions with T cells and NK cells; and a third group (TT3) primarily interacting with neutrophils. We randomly selected three samples, each dominated by the tumor cell type TT1, TT2, and TT3, respectively. Multiplex immunofluorescence (mIF) analysis was conducted on tissue sections to characterize the cellular composition.