Screening for differentially expressed metabolic genes
We downloaded the KEGG genes containing metabolic genes from the GSEA website (https://www.gsea-msigdb.org/gsea/index.jsp) Set c2.cp.kegg.v7.0.symbols.gmt, using Strawberry Perl 5.32.0 to extract a total of 921 metabolism-related genes from TCGA-COAD samples. Using R language ”Limma” software package and ”sva” software package [6], a total of 645 crossover genes were extracted from the TCGA-COAD and GSE40976 data sets and reduced the heterogeneity between the two studies. We used the Wilcox test of the R language ”Limma” software package to screen for differentially expressed genes (DEG). P Value <0.05 and |log2-fold change|> 0.5 were defined as DEG, A total of 225 DEG were screened out. The heat map and volcano map are created by the ”pheatmap” software package of the R language (https: //cran.r-project.org/web/packages/pheatmap/).