Efficiency of Bayesian quantitative trait loci mapping with full‐sib progeny

Abstract

AbstractQuantitative trait loci (QTL) mapping with perennial crops is based on one or more full‐sib progeny because homozygous genotypes are difficult to obtain. The objective of this study was to assess the efficiency of Bayesian QTL mapping with full‐sib progeny. The analyses used two simulated data sets, one assuming genotyping for 292 simple sequence repeats (SSR) loci (density of 10 cM) and the other assuming genotyping for 2969 single nucleotide polymorphisms (SNPs) (density of 1 cM). Each data set consisted of 50 replications of 40 full‐sib progeny of size 400. We assumed a broad sense heritability of 60%, genetic control by 10 QTLs and 90 minor genes, and positive dominance. The QTL heritability values ranged from 1 to 12%. The scenarios included one and multiple progeny. In the best scenarios for the low (four progeny of size 400) and high marker density (one progeny of size 400), the average power of detection was 52 and 67% for the low heritability QTLs, 83 and 95% for the average heritability QTLs, and 95 and 94% for the high heritability QTLs, respectively. The observed false discovery rate (FDR) was 15 and 9% with low and high density, respectively. The Bayesian QTL mapping provides a precise localization of candidate genes with a bias in the QTL position of approximately 4–6 cM. The polygenic effect is important to control the false discovery rate (FDR) and to provide a higher power of QTL detection with multiple progeny.