Fig. 3

Construction of differential expression of m6A gene signatures and functional annotation. A, B Functional annotation for m6A-related genes using GO enrichment analysis. The color depth of the bar plots represents the number of genes enriched. C The survival curves of the m6A phenotype-related gene signatures were estimated by the Kaplan–Meier plotter. (P < 0.001, Log-rank test). D Unsupervised clustering of overlapping m6A phenotype-related DEGs to classify patients into different genomic subtypes, termed as m6A gene Cluster A and B, respectively. The gene signature subtypes, m6A clusters, survival status, tumor stage, gender, and age were used as patient annotations. E GSVA enrichment analysis showing the activation states of biological pathways in m6A gene Cluster A and gene Cluster B modification patterns. The fraction of tumor-infiltrating lymphocytes in three m6A modification patterns. Within each group, the scattered dots represent TIME cell expression values. F ssGSEA algorithm; G CIBERSORT approach