Analysis of the Relative Effects of SNPs within a Gene to Serum Lipid Profiles Using Stepwise Linear Regression

ABSTRACT

BACKGROUND AND OBJECTIVES: It is very important to distinguish between the primary and secondary genetic effects at different sites within a small genetic region. Therefore, we evaluated the relative effects of single nucleotide polymorphisms (SNPs) within a gene on the serum lipid profiles by using individual data. SUBJECTS AND METHODS: To evaluate the contributions of SNPs in a region to the serum lipid profiles (total cholesterol, triglyceride, low density lipoprotein, high density lipoprotein), we used data that consisted of 808 individuals (327 males and 481 females) who did not have cardiovascular disease. In this study, we used a stepwise regression procedure to analyze the relative effects of four single nucleotide polymorphisms (ACE6, ACE7, ACE8, ACE10) in a gene region on the development of the serum lipid profiles in each gender group. RESULTS: In the males, there were epistatic interaction effects between two loci (ACE6xACE7, ACE6xACE8, ACE6xACE10, ACE8xACE10 and ACE7xACE8) and among three loci (ACE6xACE7xACE8, ACE6xACE7xACE10 and ACE6xACE8xACE10). Also, there are interaction effects between two loci (ACE6xACE7, ACE6xACE8, ACE6xACE10, ACE7xACE10 and ACE8xACE10) and among three loci (ACE6xACE7xACE8, ACE6xACE7xACE10, ACE6xACE8xACE10 and ACE7xACE8xACE10) in the females. CONCLUSION: The results suggested that each of these loci is important in causing a relative change of the serum lipid profiles, even with simultaneously accounting for the effects at the other loci. In the results of the analysis, there existed the effects of individual loci and significant interaction between the loci on the serum lipid profiles in each gender group. It was confirmed that this stepwise regression method can be suitable for evaluating the relative effects of SNPs and it is easily performed.