Investigating the Impact of Dominance Genetic Effects on the Accuracy of Genomic Evaluation

The review of published articles in the field of genomic selection shows that in the most of studies conducted in livestock and crop species, genomic evaluation and prediction of genomic breeding values ​​have been done without considering the dominance genetic effects. However, recent studies have shown that the dominance genetic effects contribute significantly to the phenotypic variation of productive traits of domestic animals. Therefore, it seems that ignoring the dominance genetic effects will affect the accuracy of the genomic evaluation. In this article, the consequences of not considering the dominance genetic effects in the genomic evaluation model on accuracy, mean square error, bias and reliability of genomic breeding values ​​were investigated.

A genome consisting of 5 chromosomes, each 1 Morgan length, containing 5000 bi-allelic single nucleotide polymorphism (SNP) was simulated at heritability level of 0.5. All quantitative trait loci (QTLs) were assigned additive genetic effects. Different distributions of QTL effects (uniform, normal and gamma) as well as three scenarios of the number of QTL as 5, 10 and 20% of the total number of SNPs (respectively 250, 500 and 1000 QTLs) were considered as simulation hypotheses. In different scenarios, dominance genetic effects were given to 0.00, 10, 25, 50 and 100% of QTLs. The genomic breeding values ​​were estimated using the genomic best linear unbiased prediction method (GBLUP) and the criteria of LR method such as prediction accuracy, mean square error of prediction, bias and reliability of the genomic breeding values ​​were used to analysis genomic breeding values ​​predicted by GBLUP.

The results showed that if the dominance genetic effects contribute to the phenotypic variation of the interested trait, but ignored from the genomic evaluation model and remain unseparated from the additive genetic effects, lead to a decrease in the accuracy of the genomic breeding values ​​by about 25%. Also, the mean square error of prediction of genomic breeding values ​​increased by 60% following increase in the percentage of QTLs with dominance effect from 0.00 to 100%. The bias of genomic breeding values ​​was also affected by ignoring dominance effects and increased by 36% following increase in the percentage of QTLs with dominance effect from 0.00 to 100%. Also, the reliability of genomic breeding values ​​was significantly reduced by about 40% with the increase in the percentage of QTLs with dominance effect from 0.00 to 100%.

In general, the results of this research showed that not separating the dominance genetic effects from additive genetic effects leads to low-accuracy, biased, and unreliable estimates of genomic breeding values, which will ultimately reduce the efficiency of genomic selection schemes. Therefore, it was suggested that in order to increase the efficiency of genomic selection, dominance genetic effects should be included in the genomic evaluation model.