The use of measured genotype information in the analysis of quantitative phenotypes in man

TLDR

Improved laboratory methods enable investigation of how allelic variation at a locus influences quantitative trait expression. The study introduces statistical methods that integrate measured genotype data into quantitative phenotype analysis to detect and estimate single‑locus and residual polygenic effects. The authors develop likelihoods for the joint distribution of phenotype and genotype for unrelated individuals or nuclear families and apply the model to serum cholesterol and Gc concentration. Applying the method to Gc polymorphism data, the authors simultaneously estimate genotype frequencies, genotype effects, and the contribution of unmeasured polygenes to phenotypic variability for the first time.

Abstract

Summary Improved laboratory methods allow one to investigate the contribution of measured allelic variability at a locus physiologically involved in determining the expression of a quantitative trait. We present statistical methods that incorporate measured genotype information into the analysis of a quantitative phenotype that allows one simultaneously to detect and estimate the effects of a measured single locus and residual polygenic effects. Likelihoods are presented for the joint distribution of the quantitative phenotype and a measured genotype that are appropriate when the data are collected as a sample of unrelated individuals or as a sample of nuclear families. Application of this method to the analysis of serum cholesterol levels and the concentration of the group specific component (Gc) are presented. The analysis of the contribution of the common Gc polymorphism to the determination of quantitative variability in Gc using smaples of related and unrelated individuals presents, for the first time, the simultaneous estimation of the frequencies and the effects of the genotypes at a measured locus, and the contribution of residual unmeasured polygenes to phenotypic variability.