A Novel Association Approach To Search For Coronary Artery Disease Genetic Related Metabolites Based On SNPs And Metabolic Network

Coronary artery disease (CAD) is rapidly becoming the leading cause of death inthe world, considered to be closely relevant with genesis and biological metabolism.Current studies on CAD are basically relying on detecting the variation of CADrelated metabolic bio-marker levels and identifying their functions for clinicaldiagnosis and treatment. Understanding the role of metabolic compounds and theirinteraction with environmental factors in CAD is key to the development of safe andefficient therapies, to diagnosis and to prevention.Metabolic compounds as the ultimate product during cell regulation process, itslevel can reflect how genetic factors and surrounding circumstance influence onbiological processes. The incidence of CAD is often accompanied by metabolites’changes in metabolic processes. In the level of cellular biology, there has been founda variety of mechanisms and metabolic pathways associated with CAD. For example,mitochondria oxidative metabolism and ATP production, lipoprotein metabolism,oxidative stress. To identify CAD related metabolites are increasingly needed in thepre-clinical and clinical diagnosis for better evaluation of novel chemical entities andtarget validation.The genetic predilection of CAD is well established, however, family history hasbeen proved to be one of the independent risk factors for CAD, especially inearly-oneset forms. Many commonly accepted risk factors for CAD need furtherexploration. Gene mutation is proved to be associated with many risk factors such assingle nucleotide polymorphism (SNP). In previous studies, scientists utilized humanmetabolic pathways and high-throughput SNPs datasets to find genetic risk pathwayswhich are related with complex diseases. The research successfully utilized SNPs dataand built a scoring measurement which has obtained well effects.Metabolomics, the study of small-molecules in cells, tissues and biological fluids,has become increasingly used for diagnosis of human disease. Studies in myocardial ischemia using unbiased metabolomic profiling have uncovered differences in profiles.But most analytes within the profile remained unidentified, making clinical andmechanistic interpretation difficult. However, with the development of metabolicnetwork, more and more scientists use metabolic network as one of the effect ways inmetabolic research. For example, Hojung Na et al. adopted the enzyme activity dataand gene expression data profiles analyzed bacillus coli metabolic network by usingan improved Z-score method. Combining metabolic network with metabolites riskevaluation may help to further understanding the pathogenesis of CAD in both geneticand metabolic level.The Edinburgh Human Metabolic Network (EHMN) has been widely used incomplex disease, it was created with a largely automated top-down approach. EHMNhas eight compartments (cytoplasm, extracellular space, mitochondria, Golgiapparatus, endoplasmic reticulum, lysosome, peroxisome and nucleus). Recently, etproposed a new computational approach that can identify disease biomarkers whichthey called reportor metabolites in EHMN.Herewith, in this study, we proposed a computational approach (RCM) to identifyCAD significant genetic risk compounds (SGRCs) based on EHMN andhigh-throughput SNPs datasets. As it is model based, the current approach applies toeight sub-cellular networks, respectively. After finding CAD genetic risk compounds,we adopted CAD related genes to validate the feasibility and effectively of ourapproach. Besides, some of the SGRCs are confirmed to be related with CADpathogenesis by literature search and functional classification. These results providethe evidence that the high heritabilities of metabolites play an important role in CADpathogenesis and the newly discovered metabolites could possibly serve asbiomarkers for CAD clinical diagnosis.