Statistical Tests and Threshold Values for Locating Quantitative
           Trait Loci (QTL) in Experimental Populations


                       R. W. Doerge
                   Department of Statistics
                     Purdue University




%The detection and location of genes that are responsible for
%quantitative characters is a problem of great interest to the genetic
%mapping community.  Interval mapping has proved to be a useful tool in
%locating quantitative trait loci (QTL), but has recently been challenged by
%faster, more sophisticated regression methods.  Regardless of the method
%used to locate QTL, the distribution of the test statistic (LOD score
%or likelihood ratio test) is questionable.  Due to the
%quantitative trait value distribution following a mixture distribution
%rather than a single distribution,  the asymptotic distribution
%of the test statistic is not from a standard family, such as chi-square.
%The purpose of this work is to review interval mapping, discuss the
%distribution of the resulting test statistic, and then
%present empirical threshold values for the declaration of major QTL,
%as well as minor QTL.  Empirical threshold values are obtained by
%permuting the actual experimental trait data, under a fixed and known genetic map, for the
%purpose of representing the distribution of the
%test statistic under the null hypothesis of no QTL effect.  Not only
%is a permutation test statistically justified in this case, it accurately
%retains the specifics of the experimental situation under investigation
%(i.e., sample size, marker density, skewing, etc.), and may be used
%in a conditional sense to derive thresholds for minor QTL once
%a major effect has been determined.
%