Fig. 4

Construction of gene clusters based on PACs-related DEGs. A Identification of three gene clusters (k = 3) based on PACs-related DEGs by consensus clustering algorithm, gene cluster A (534 patients), gene cluster B (465 patients), and gene cluster C (335 patients). B The remarkable difference among transcriptome profiles of three gene clusters with PCA. C The OS of three gene clusters in TCGA and five GEO cohorts shown by Kaplan-Meier curves (log-rank tests, P < 0.001). D Relationships between clinicopathologic features and three gene clusters analyzed by unsupervised clustering. E Differences in the expression of 49 PYAGs among three gene clusters. F Differences in the expression levels of immune checkpoints among three gene clusters in TCGA and five GEO cohorts. *P < 0.05; **P < 0.01; ***P < 0.001. DEGs: Differentially expressed genes; PCA: Principal component analysis; OS: Overall survival