Abstract
Inbreeding depression is a highly relevant universal phenomenon in population and conservation genetics since it leads to a decline in the fitness of individuals. This phenomenon is due to the homozygous expression of alleles whose effects are hidden in heterozygotes (inbreeding load). The rate of inbreeding depression for quantitative traits can be quantified if the coefficient of inbreeding (F) of individuals is known. This coefficient can be estimated from pedigrees or from the information of molecular markers, such as SNPs, using measures of homozygosity of individual markers or runs of homozygosity (ROH) across the genome. Several studies have investigated the accuracy of different F measures to estimate inbreeding depression, but always assuming that this is only due to recessive or partially recessive deleterious mutations. It is possible, though, that part of the inbreeding depression is due to variants with overdominant gene action (heterozygote advantage). In this study, we carried out computer simulations to assess the impact of overdominance on the estimation of inbreeding depression based on different measures of F. The results indicate that the estimators based on ROH provide the most robust estimates of inbreeding depression when this is due to overdominant loci. The estimators that use measures of homozygosity from individual markers may provide estimates with substantial biases, depending on whether or not low-frequency alleles are discarded in the analyses; but among these SNP-by-SNP measures, those based on the correlation between uniting gametes are generally the most reliable.