%0 Generic
%J Proceedings of the 8th World Congress on Genetics Applied to Livestock Production, Belo Horizonte, Minas Gerais, Brazil, 13-18 August, 2006
%D 2006
%T Statistical power and sample size requirement for QTL detection.
%A London, N. R.
%A Da, Y.
%K animal breeding
%K backcrosses
%K dominance
%K gene mapping
%K genetic effects
%K genetic parameters
%K heritability
%K quantitative trait loci
%K statistical analysis
%X Statistical power and sample size requirements for detecting additive and dominance effects under the F-2 and reciprocal backcross (RBC) designs were derived as functions of a set of genetic and statistical parameters. The resulting formulae for statistical power are increasing functions of sample size, type-I error, and additive and dominance effects, and are decreasing functions of marker-QTL recombination frequencies. The behavior of statistical power for the F-2 and RBC designs was studied using the exact mathematical expressions and Monte Carlo simulations. The theoretical and simulation results agreed well. The F-2 design is more powerful for detecting additive effects, whereas the RBC design is more powerful for detecting dominance effects. The differences in statistical power are most pronounced for low heritabilities, small sample sizes, and large recombination frequencies. Sample size requirements were also derived as functions of a number of statistical and genetical parameters. With these results, statistical power and sample size requirement of a QTL mapping experiment can be predicted for any given set of statistical and genetical parameters.
%B Proceedings of the 8th World Congress on Genetics Applied to Livestock Production, Belo Horizonte, Minas Gerais, Brazil, 13-18 August, 2006
%P 22.18
%G eng