A simple correction for multiple comparisons in interval mapping genome scans

Several approaches have been proposed to correct point-wise significance th resholds used in interval-mapping genome scans. A method for significance t hreshold correction based on the Bonferroni test is presented. This test in volves calculating the effective number of independent comparisons performe d in a genome scan from the variance of the eigenvalues of the observed mar ker correlation matrix. The more highly correlated the markers, the higher the variance of the eigenvalues and the lower the number of independent tes ts performed on a chromosome. This approach was evaluated by mapping 1000 n ormally distributed phenotypes along chromosomes of varying length and mark er density in a population size of 500. Experiment-wise significance thresh olds obtained from the simulation are compared to those calculated using th e Bonferroni criterion and the newly developed measure of the effective num ber of independent tests in a genome scan. The Bonferroni calculation produ ced significance thresholds very similar to those obtained by simulation. T he threshold levels for both Bonferroni and simulation analysis depended st rongly on the marker density and size of chromosomes. There was a slight bi as of about 1% in the thresholds obtained at the 5% and 10% point-wise sign ificance levels. The method introduced here provides a relatively simple co rrection for multiple comparisons that can be easily calculated using stand ard statistics packages.