Fig. 4

Consensus clustering analysis according to DEGs of m6Aclusters. A DEGs were analyzed in every pairwise comparison of m6Aclusters and intersecting genes were identified via a Venn diagram. B Univariate Cox regression analysis was performed to evaluate RFS in relation to 74 intersecting genes. The statistical significance (P < 0.05) of 6 genes (BAIAP2, TEX264, MMAB, JAGN1, TIMM8AP1, and IMP3), P value, and HR value (with 95% CI) are shown in the plot. C Consensus clustering analysis according to DEGs of m6Aclusters was performed. (left) Relative change in area under the CDF curve from k = 2 to 9. (right) Color-coded heatmap of the consensus matrix for k = 3. Color gradients represent values from 0–1 (white: 0, dark blue: 1). D The heatmap shows unsupervised clustering of the 6 genes in TCGA-PRAD. PSA grade, pT, pN, GS, and biochemical recurrence were used for patient annotation. E Survival analysis for RFS among three geneClusters based on 495 TCGA-PRAD patients. Kaplan–Meier curves and log-rank P values are shown in the graph, and the numbers of patients at risk are shown at the bottom. F Expression of m6A regulators among three geneClusters is shown in the boxplot. The median ± interquartile range of values is shown in the graph. ns P > 0.05; * P < 0.05; ** P < 0.01; ***P < 0.001. G As described in Fig. 2F, intersecting DEGs of GS, pT, RFS, TN, and TP53 are shown in a Venn diagram. H As described in Fig. 2G, PCa tissues and adjacent normal tissues were divided into four groups, and mRNA levels of intersecting gene (MMAB) in different groups were compared. * P < 0.05