Construction and verification of risk scoring prognostic model
After 1000 resampling, 22 metabolic genes were subjected to LASSO Cox regression analysis to construct a prognostic model, which containing 16 metabolic genes. It shows the 16 genes and their coefficients in the risk scoring model (Table 1 ). Calculate the risk score of each patient based on the mRNA expression level and risk coefficient of each gene.We divided the TCGA-COAD and GSE40976 samples into high-risk groups and low-risk groups based on the median risk score. Kaplan–Meier analysis was performed to prove that the overall survival of the high-risk group was poor (Figure 2A, B). The risk score distribution showed that the mortality rate of the high-risk group was higher than that of the low-risk group (Figure 2C, D). A heat map was developed to show the high-risk and low-risk TCGA-COAD and GSE40976 gene expression profiles (Fig. 2E, F). The heat map shows the expression of 16 gene markers. SEPHS1, P4HA1, ENPP2, PTGDS, GPX3, CP, ASPA, POLR3A, PKM and POLR2D are positively correlated with high-risk groups, indicating that high expression of these genes is associated with a shorter overall survival time. XDH, EPHX2, ADH1B, HMGCL, GPD1L, and MAOA revealed opposite effects, indicating that high expression of these genes is associated with longer overall survival time. P Value<0.05 is considered statistically different.