[Retracted] Characterization of Tumor Mutation Burden-Based Gene Signature and Molecular Subtypes to Assist Precision Treatment in Gastric Cancer

Figure 4

A LASSO model based on TMB-related DEGs links to gastric cancer patients’ OS in TCGA cohort. (a) Association between log and partial likelihood deviance. 10-fold cross-verification was used for selecting the optimal parameter in the LASSO model. (b) LASSO regression coefficients for genes in the model. (c) TMB risk score distribution. Vertical dotted line represented the risk score median. Patients were stratified into high- and low-risk groups. Red was indicative of high-risk group and green was indicative of low-risk group. (d) Survival status (green: alive; red: dead) distribution. (e) OS probabilities for high- and low-risk groups. value was assessed via log-rank test. (f) Heatmap for expression profiles of 4 genes in this model. (g) ROC curve of OS in TCGA cohort.

(a)

(b)

(c)

(d)

(e)

(f)

(g)