Development of a Novel KCNN4-Related ceRNA Network and Prognostic Model for Renal Clear Cell Carcinoma

Figure 7

Establishment of the ce-RNA based-gene signature and prognosis model. (a) The heatmap of KCNN4-centred ce-RNA network. (b) Univariate Cox regression analysis of all genes. -Values and HRs were shown. (c) and (d) LASSO regression analysis of all lncRNAs to pick out applicable ones for gene signature. (e) Univariate Cox regression analysis of selected four lncRNAs, together with KCNN4 and miRNA in 514 ccRCC patients from TCGA database. (f) Survival analysis of ccRCC patients from high- and low-risk groups according to the risk scores from gene signature. (g) Nomogram of the prognosis model derived from the risk signature. Survival rate at 1-, 3-, and 5-year from the model was shown as well. (h) ROC curves to validate the sensitivity and specificity of the model. AUC at 1, 3, and 5 years was noted. (i) Calibration curve to validate the predictive efficiency of the model. Y-axis represents the actual 3-year OS and X-axis represents the predicted 3-year OS. Abbreviations: ROC, receiver operating characteristic; AUC, area under curve.

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(i)