Estimating QTL location and QTL variance in half-sib families under the random model with missing parental genotypes

A sib-pair analysis was carried out to estimate the location and the variance of the quantitative trait locus (QTL) in half-sib families using an interval mapping procedure based on a random model. The (co)variances between half-sibs were modelled as a function of the proportion of alleles identical-by-descent (IBD) shared by two half-sibs at the locus affecting the trait. The proportion of alleles IBD at the QTL was estimated from the IBD at two flanking marker loci, using half-sib and parental information. With missing parental marker genotypes the IBD proportion was approximated. Data were generated under a monogenic model, varying heritability of the trait, number of families, and heterozygosity of marker loci. QTL position and variance were estimated with a QTL simulated terminally or centrally on the chromosome, for situations when both parental genotypes are known, and when one or both parental genotypes are missing. With lour heritability, the estimated position of the terminally located QTL was biased toward the middle of the chromosome. With high heritability, the QTL variance was underestimated when both parental genotypes were known. Better estimates for QTL variance were obtained with unknown parental genotypes, because the number of non-informative sib-pairs was reduced when the IBD proportion at marker loci was approximated. Less biased and more accurate estimates and higher power of QTL detection were obtained with higher heritability and a larger number of families. Marker heterozygosity did not influence estimation of QTL position. Increased marker heterozygosity resulted in overestimation of the polygenic variance with missing parental genotypes.