Fig. 4
Application of PACSI on BRCA data. A The UMAP visualization of the BRCA scRNA-seq dataset. B The UMAP visualization of PACSI-identified cells. The red dots are PACSI-identified cells associated with poor prognosis in BRCA. C The distribution of PACSI-identified cells by cell clusters. D Rank for regulons in PACSI-identified cells associated with poor prognosis in BRCA based on RSS. E The differentially expressed genes (log2 fold change > 2, FDR < 0.01) in PACSI-identified cells (labeled 1) and all the other cells (labeled 0). F Violin plots show the expression levels of vital genes in PACSI-identified cells (n = 317) and all the other cells (n = 1217). Two-tailed P value was calculated by Wilcoxon rank-sum test. G The top five Reactome pathways enrichment of genes that were expressed higher in the PACSI-identified cells. H Kaplan–Meier estimates survival curves for the high-risk and low-risk groups according to PACSI-derived signature. The log-rank test was used to calculate P values. The x-axis indicates the follow-up time; the y-axis indicates the probability of overall survival (OS), disease-free survival (DFS), and recurrence-free survival (RFS)