Figure 2. Construction of molecular subtypes and prognostic analysis. (A) The differences of DNA sensors expression between tumor and adjacent tissues. (B) Venn diagram revealing 30 genes in relation to prognosis of HCC by univariate COX analysis. (C) CDF curves and CDF delta area curves. The optimal number of clusters is the K value where the CDF curve is the smoothest and also the turning point of the Delta area. (D) Subtype classification of HCC samples. (E) Prognostic analysis in TCGA dataset. (F) Prognostic analysis in HCCDB18 dataset. (G) Prognostic analysis in GSE76427 dataset. The line in the box represents the median value, and the asterisks represent the P-value (***p < 0.001; ****p < 0.0001). The statistical analyses were performed by the student t-test. HCC, hepatocellular carcinoma; CDF, cumulative distribution function; ns, not significant.