Figure 2.
Enrichment analysis and consensus clustering for determining the optimal K-value and validating the immune typing strategy. (A, B) Gene Set Enrichment Analysis (GSEA) of signaling pathways enriched in non-recurrence vs. recurrence subgroups. (C) Consensus matrix (CM) plots at K = 3, showing optimal cluster separation. (D) Cumulative distribution function (CDF) plots for K-values ranging from 2 to 10. (E) Area under the CDF curve, indicating the stability of clustering at K = 3. (F) Average consistency analysis, confirming K = 3 as the most stable clustering solution. (G) Heatmap of tumor microenvironment (TME)-related scores, including tumor purity, ESTIMATE, immune, and stromal scores. (H) Comparisons of TME-related scores across different immune subgroups. (I) Boxplot of HLA gene expression levels across immune subgroups. ***p < 0.001.
