An application of a model for a genotype-dependent relationship between a concomitant (age) and a quantitative trait...

In most genetic studies in humans the variability in a quantitative trait is adjusted for variability in concomitants (age, sex, etc) using a single regression equation prior to analyses of pedigree data. To illustrate an alternative approach, a single locus genetic model was tested. This model incorporates genotypic effects on the level of the trait, the variability in the trait, and the relationship between a concomitant and the trait. In this study, the model was applied to measures of age and low-density lipoprotein (LDL) cholesterol in a large kindred with familial hypercholesterolemia. The application of this model to 322 individuals in four generations provided evidence that genotypic variation at a single locus influences LDL levels early in life, the rate of increase of LDL with age and the phenotypic variance. A model with genotype-dependent slope and variance fit the data significantly better than a model with slope and variance independent of genotype. The inclusion of age-specific genotypic differences contributed to identification of high-risk individuals, to statistical support for a major locus, and to evidence for genetic determination of the tracking of LDL levels. Models that incorporate genotype-specific concomitant effects have the potential to represent more realistically the relationship between genotypic variability and quantitative phenotypic variation than models that assume that these effects do not exist.