Author/Authors :
Mehrabi، Yadollah نويسنده , , SARBAKHSH، PARVIN نويسنده Department. of Statistics and Epidemiology, School of Public Health, Tabriz University of Medical Sciences, Tabriz, Iran , , J. Houwing-Duistermaat، Jeanine نويسنده Department of Medical Statistics and Bioinformatics, Leiden University Medical Centre, Leiden, The Netherlands , , Zayeri، Farid نويسنده Proteomics Research Center, Faculty of Paramedical Sciences, Shahid Beheshti University of Medical Sciences, Tehran , , Sadat Daneshpour، Maryam نويسنده Cellular and Molecular Endocrine Research Center, Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, IR Iran ,
Abstract :
Serum total cholesterol is an established risk factor for coronary heart diseases. In addition to environmental factors and main effects of polymorphisms, genetic interactions can influence cholesterol level. Furthermore, polymorphisms effects can be time-dependent. So, they could be valuable to study of genetic interactions over time. In this study we proposed logic mixed model to assess the association of Single-nucleotide polymorphism and other risk factors with longitudinal quantitative cholesterol level with respect to their interactions. Data of 329 subjects with age ≥ 20 years that participated in Tehran Lipid and Glucose Study with complete data for three phases of study was analyzed using proposed model. The results showed that the cholesterol level of male or subjects with GG genotype for ApoAIV and normal waist circumstance and normal blood pressure was 19.8 mg/dL less than other subjects without this combination (β = -19.8, CI 95%: -13.9, -25.69). Also, having "high triglyceride or allele e2 for ApoE" was associated with an increased effect on cholesterol (β = 16.1, CI 95%: 11.64, 20.55). The level of the cholesterol in phase one of study was 21.4 mg/dL (CI 95%: 18.35, 24.44) more than other phases. The variance component of the random effect was statistically differing with zero. In this study, we extended logic regression to the longitudinal quantitative response and applied it to the TLGS data. We identified some interactions among SNPs and other covariates related to cholesterol level using logic mixed model.
