Systematic analysis of coronary artery disease datasets revealed the potential biomarker and treatment target

Coronary artery disease caused about 1 of every 7 deaths in the United States and early prevention was potential to decrease the incidence and mortality. We aimed to figure the genes involving in the coronary artery disease using meta-anlaysis. Five datasets of coronary heart disease from GEO series were retrieved and data preprocessing and quality control were carried out. Moderated t-test was used to decide the differentially expressed genes for a single dataset. And the combined p-value using systematic-analysis methods were conducted using MetaDE. The pathway enrichment was carried out using Reactome database. Protein-protein interactions of the identified differentially expressed genes were also analyzed using STRING v10.0 online tool. After removing unidentified or intermediate samples and a total of 238 cases and 189 matched or partially matched control from five microarray datasets were retrieved from GEO. Six different quality control measures were calculated and PCA biplots were plotted in order to visualize the quantitative measure. The first two PCs captured 91% of the variance and we decided to include all of the datasets for systematic analysis. Using the FDR cut-off as 0.1, nine genes, including LFNG, ID3, PLA2G7, FOLR3, PADI4, ARG1, IL1R2, NFIL3 and MGAM, were differentially expressed according to maxP. Their protein-protein interactions showed that they were closely connected and 24 Reactome pathways were related to coronary artery disease. We concluded that pathways related to immune responses, especially neutrophil degranulation, were associated with coronary heart disease.