Bivariate variance component analyses of genetic and shared environmental correlations – with application to blood pressure and body mass index
by
Jisheng Cui
The University of Melbourne
Coauthors: John L Hopper, Stephen B Harrap

Genes and shared living environments may concurrently determine the correlations between two or more phenotypes of an individual and among family members. Understanding such genetic and shared environmental correlations is important to postulate efficient and effective strategies to prevent and reduce the risks of cardiovascular disease. We applied a bivariate variance component analysis method to estimate the genetic and shared environmental variance components and their correlations between two phenotypes under the assumption of multivariate normal distribution. We also used an independent univariate analysis approach to investigate changes of the variance components of one phenotype when adjusted for another. The calculation was carried out using specialist software FISHER. We applied this method to blood pressure (BP) and body mass index (BMI) of 767 adult two-generation families (2912 individuals) from the Victorian Family Heart Study. For systolic blood pressure (SBP) and diastolic blood pressure (DBP), the correlation between the effect of genes was estimated to be 1.0 and between shared environment was 0.52 (s.e. 0.1). For SBP and BMI, the correlations were 0.30 (s.e. 0.1) for genetic and 0.22 (s.e. 0.09) for shared environmental effects, respectively. Our findings suggest that the same genes and many shared environmental factors determine both SBP and DBP. In contrast, SBP and BMI share genetic and environmental determinants to a lesser degree.

Date received: April 4, 2002