Results from Simulations to Show the Relative Impact of Uninformative Pairs on QTL Analysis for Different Values of Polygenic and QTL Heritability and Different Percentages of Uninformative Markers HERITABILITY

a The data in the first two rows correspond to the simulated scenario (2) of Cordell (2004) and are based on 10,000 replicates. For all other data, a normally distributed additive QTL was simulated, and results are averages from 1,000 replicates. b dT is the average difference between the test statistic achieved when uninformative pairs are removed from the analysis and the one achieved when they are kept in the analysis. j(d) is the SD of the difference between the test statistic achieved when uninformative pairs are removed from the analysis and the one achieved when they are kept in the analysis. E(T) is the average test statistic achieved when uninformative markers are removed from the analysis. j(T) is the SD of the test statistic when uninformative markers are removed from the analysis. All ratios are expressed as percentages.