Using Gene Expression Profile to Extract the Biomarker Genes of Cardiovascular Disease

Abstract

Cardiovascular disease (CVD) is the world’s premier cause of morbidity and death. CVD is a class of heart or blood vessel diseases. CVD contains the coronary artery disease (CAD) such as unstable angina (UA) and myocardial infarction (MI) diseases. Clinicians use additional tools to support clinical evaluation and improve their ability to detect the susceptible patient at threat for CVD. Biomarkers are one such method to identify potential risk persons, rapidly and reliably diagnose disease symptoms that efficiently predict and treat disease. Discovery of MicroRNAs (miRNAs) representing a class of small, non-coding RNA molecules opens interesting opportunities to use the patterns of miRNAs as a biomarker for cardiovascular diseases. The objective of this study is to define miRNA and genes potentially associated with MI. Rothman dataset includes 52 samples of Acute Coronary Syndromes (ACS). including 18 patients with myocardial infarction (MI) and 8 patients with unstable angina (UA). Overall (number of genes selected) candidate ncRNA biomarkers have been defined and a ncRNA-based classifier has been created to predict MI risk which based on 7 ncRNA expression data using vector support machines SVM and decision tree classifiers. The experimental results obtained through applying these mechanisms on the Rothman dataset. The classification model’s performance is evaluated using the V-fold validation and LOOCV methods. The outcome of this search can be used by the drug designer for pathway analysis and CVD treatment decisions.