Extended Data Fig. 6: Spatial organization and cell-cell interactions of CMs.
From: Cross-tissue multicellular coordination and its rewiring in cancer
(a) Schematic of two spatial metrics for CMs using 10x Visium data. The colocalization score evaluates the extent to which two cell subsets colocalize within the same spots, while the aggregation score measures the spatial adjacency of two cell subsets, encompassing both colocalization and broader spatial relationships. (b) Spatial aggregation (top) and cellular composition (bottom) of CMs. Top: CM aggregation scores across Visium samples colored by tissues (Methods). CMs are ordered by the median of aggregation scores across samples. Bottom: Proportion of different cell lineages in all cell-subset nodes within individual CM networks shown in Fig. 2b. None = 0 <Low <0.2 ≤ Medium <0.4 ≤ High. (c,d) CellPhoneDB cell-cell interaction analysis based on scRNA-seq data. Heat maps showing of cell-cell interaction counts within and across cell types (c) or CMs (d). Bar plots on the top and right represent the sums of the columns and rows, respectively. (e) CellChat cell-cell interaction counts based on scRNA-seq, stratified by ligand-receptor pair types. (f) Heat map showing the responsive cytokines (columns) among cell subsets within individual CMs (rows) (Methods). Bar plots on the top and left represent the sums of columns and rows, respectively. (g) Cytokine responsiveness in CMs. For each CM, the CM FDR denotes the minimum FDR value among all subsets within the CM (Methods). For box plot in b, the center line represents the median, the box limits delineate the upper and lower quartiles, and the whiskers extend to the highest and lowest values within 1.5× the interquartile range.
