Identification of 5-Gene Signature Improves Lung Adenocarcinoma Prognostic Stratification Based on Differential Expression Invasion Genes of Molecular Subtypes

Figure 5

Construction and verification of prognostic models based on differential genes of molecular subtypes: (a, b) the distribution of RiskScore and the corresponding distribution of survival state in TCGA training data set; (c) heat map of gene expression of the RiskScore model; (d, e) KM curve and ROC curve of the high- and low-risk group in TCGA training data set; (f, g) KM curve and ROC curve of the high- and low-risk group in TCGA validation data set; (h, i) KM curve and ROC curve of the high- and low-risk group in all TCGA data sets.

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(i)