Comparisons of quantitative trait locus mapping properties between two methods of recombinant inbred line development

Abstract

Summary Theoretical comparisons for quantitative trait loci (QTL) mapping properties were conducted among simulated recombinant inbred (RI) populations developed by single-hill (SH), complete bulk, and single seed descent (SSD) procedures by Monte Carlo simulations based on various population sizes, heritabilities, and QTL effects. Our simulations included estimation of QTL effects, QTL positions, and statistical testing power in the RI populations by comparing the estimates with preset values. The simulation results showed that the single hill (SH) bulk and single seed descent RI populations were generally not significantly different with respect to quality of estimated QTL effects and positions. Furthermore, when each RI population had 150 lines, each could provide desirable properties for QTL mapping. The results implied that a SH RI population consisting of 75 or more F2-derived families with two lines per family (corresponding population size of 150 or above) was appropriate for QTL mapping and was not significantly different than a SSD RI population of 150. Thus, the SH method could be used to develop large numbers of RI lines for achieving better results in QTL mapping. Simulations also showed that there was no significant difference between means using SH methods with 10 and 100 fruits per family. However, RI populations developed by the complete bulk method where F2 identities are lost were not suitable for QTL mapping.