Integrated Analysis of Gene Expression and Methylation Data to Identify Potential Biomarkers Related to Atherosclerosis Onset

Abstract

Atherosclerosis is a kind of chronic inflammatory cardiovascular disease. Epigenetic regulation plays a crucial role in atherosclerosis. Our study was aimed at finding potential biomarkers associated with the occurrence of atherosclerosis. Two datasets were downloaded from the Gene Expression Omnibus (GEO) database. The epigenome-wide association study (EWAS) analysis was performed on methylation data using CpGassoc package. The differential expression analysis was conducted on mRNA data using limma package. The GO (Gene Ontology) and KEGG (Kyoto Encyclopedia of Genes and Genomes) functional enrichment was done in clusterProfiler package. Finally, the logistic regression model was constructed using generalized linear model (glm) function. Between atherosclerotic vs. nonatherosclerotic samples, totally 4980 cytosine-phosphate-guanine (CpG) sites (annotated to 2860 genes) and 132 differentially expressed genes (DEGs) related to atherosclerosis were identified. The annotated 2860 genes and 132 DEGs were significantly enriched in 9 and 4 KEGG pathways and 289 and 132 GO terms, respectively. After cross-analysis, 6 crucial CpG sites were screened to build the model, including cg01187920, cg03422911, cg08018825, cg10967350, cg14473924, and cg25313204. The diagnostic model could reliably separate the atherosclerosis samples from nonatherosclerotic samples. In conclusion, the 6 CpG sites are probably potential diagnostic biomarkers for atherosclerosis, including cg01187920, cg03422911, cg08018825, cg10967350, cg14473924, and cg25313204.