EXPERIMENTAL DESIGN AND SAMPLE SIZE REQUIREMENT FOR QTL MAPPING

Bioinformatics Research Center, Departments of Statistics and Genetics, North Carolina State University, Raleigh NC 27695-7566

There are a number of factors that influence design of a QTL mapping experiment, such as purpose of study, organism used, population cross, and of course resource available. The calculation of sample size required to detect a QTL depends on the overall heritability of a trait in the cross population, possible linkage of multiple QTL and also statistical method used QTL mapping analysis. This talk will discuss the genetic and statistical basis used for power calculation and sample requirement for QTL detection in the setting of composite interval mapping and multiple interval mapping analysis. Questions discussed include: How large sample size is required to detect a QTL that explains x percentage of the variance for a trait with y heritability with z power at the usual requirement of genome-wide significance level? How large sample size is required to separate two QTL that are, say, 15 cM apart?