Fig. 6: Proteomic stratification of the CC cohort and the corresponding protein pathways and subgroup-specific kinases.

a Patient subgrouping analysis of proteomic profiling identified three proteomic subgroups: Subgroup 1 (cyan, n = 53), Subgroup 2 (blue, n = 61), Subgroup 3 (red, n = 22). Each column represented a tumor sample, the associations of proteomic subgroups with clinical characteristics and somatic mutations were annotated in the top panel (Chi-square test or one-way ANOVA test, BH adjusted P). Rows in the bottom panel indicated differentially expressed proteins (One-way ANOVA test, Bonferroni-adjusted P < 0.05), color of each cell showed TMT protein abundance. b Kaplan-Meier curves of PFS for radical radiotherapy patients in each TMT proteomic subgroup (two-sided log-rank test). The comparison of immune score (c) and tumor size (d) among TMT proteomic subgroups (two-sided Student’s t test). Boxplots in c show the median (central line), the 25–75% IQR (box limits), and the ± 1.5 × IQR (whiskers). Data in d are represented as mean ± SEM. e Transcriptome-based deconvolution of mRNA transcript cell signatures in tumors using Xcell (one-way ANOVA test, BH adjusted P). f Differential phosphorylation sites among three TMT proteomic subgroups (one-way ANOVA test, Bonferroni-adjusted P < 0.05). Unsupervised clustering and biological pathways significantly enriched are presented on the right (one-sided Fisher’s exact test, BH adjusted P < 0.05). g Circular plot depicting the active kinases in each TMT proteomic subgroup compared to all other subgroups identified by KSEA (see Methods). The eight kinase-regulated phosphorylation sites with the highest t-statistics were indicated by black dots. h Heatmap showing the protein (TMT and DIA) and phosphorylation abundance of TMT subgroup-specific kinases. Source data are provided as a Source Data file.