Screening of miRNA target genes in coronary artery disease by variational Bayesian Gaussian mixture model

Coronary artery disease (CAD) is a leading cause of death, and microRNAs (miRNAs) are widely involved in physiological and pathological processes of CAD. We chose the targetscore method calculated via the variational Bayesian Gaussian mixture model (VB-GMM) as the prediction method of target genes. By observing the density overlap, we selected the thresholds of miRNA-1 and miRNA-155. In total, 18 target genes of miRNA-1, and 19 target genes of miRNA-155 were identified. The threshold of miRNA-146a was selected using the |logFC| value, and 16 target genes were screened out. In this study, our major contribution was to predict the target messenger RNAs (mRNAs) of the chosen miRNAs with the gene expression profiles, which can effectively reduce the workload of screening. Although the validated genes constituted only a small part in the final prediction results, it is a good sign for research in the future. It means that we could provide new research aims for future studies focusing on miRNA regulatory mechanisms.