Atherosclerosis diagnosis model based on autophagy related genes and animal experiment verification

WANG Zhu; HE Chuan-hui; YANG Hui-yu

doi:10.3969/j.issn.1672-5301.2025.10.014

Abstract：Objective To establish and validate a diagnostic model of atherosclerosis(AS)based on autophagy related genes.Methods The GSE100927 and GSE28829 datasets were downloaded from Gene Expression Omnibus(GEO)database( large amount of autophagy related gene expression data were obtained from the Human Autophagy Database(HADb)( the genes related to AS were screened.By analyzing the intersection of autophagy related genes and differentially expressed genes in the GSE100927 dataset,a total of 19 autophagy related genes were screened for differential expression.Subsequently,the machine learning methods were employed and validated using the GSE28829 dataset,ultimately six key genes identified.A new diagnostic model for AS was constructed based on these six genes and the corresponding nomogram was generated.Results The new diagnostic model based on NCKAP1,ATG16L2,CCR2,HSPB8,CTSD and RGS19 showed good accuracy and sensitivity in the diagnosis of AS.In addition,the roles of these 6 genes in 28 types of immune cells were explored through immune infiltration analysis.Further in vivo experimental verification showed that compared with the control group,the relative mRNA expression levels of ATG16L2,CCR2,CTSD and RGS19 in the AS group were significantly increased(all P<0.05),while the relative expression levels of NCKAP1 and HSPB8 were significantly decreased(all P<0.05).Conclusion The AS diagnostic model based on autophagy related genes is successfully established,providing new ideas and methods for early diagnosis and treatment of AS.