Abstract
BACKGROUND: Lipid metabolism and regulated cell death (RCD) play a role in the remodeling of tumor immune microenvironment and regulation of cancer progression. Since the underlying immune mechanisms of colon cancer remain elusive, this study aims to identify potential therapeutic target genes. METHODS: Differential genes related to lipid metabolism and RCD in COAD patients were identified using R language and online tools. Based on the expression of genes, two groups were classified using consensus clustering. CIBERSORT and ssGSEA were used to detect immune infiltration in both groups. Prognostic signature genes for colon cancer were screened using machine learning algorithms. KEGG, GO and GSEA for gene pathway enrichment. In addition, interacting genes in the immune module were obtained using a weighted gene co-expression network (WGCNA). Finally, expression and mutation of key in colon cancer genes were detected using TIMER, HPR, cBioPortal website and qPCR. RESULTS: The consensus clustering analysis revealed that 231 relevant differential genes were highly associated with immune infiltration. A series of machine learning and website analyses identified AGT as a hub gene linked to lipid metabolism and regulated cell death, which is overexpressed in colon cancer. CONCLUSION: AGT, as a signature gene of lipid metabolism and regulated cell death, plays a critical role in the development of COAD and is associated with tumor immune infiltration.