Strategies to improve on selection based on estimated breeding values

Selection of individuals based on their estimated breeding values (EBV) aims to maximize response to selection in the next generation under an additive model. However, when the aim does not only include short-term population-wide genetic gain but also genetic gain over multiple generations, an optimal strategy is not as clear-cut, as maintenance of genetic diversity may become an important factor. This study provides an extended comparison of existing selection strategies in a unifying testing pipeline using the simulation software MoBPS.

Applying a weighting factor on estimated SNP effects based on the frequency of the beneficial allele resulted in an increase of the long-term genetic gain of 1.6% after 50 generations, while reducing inbreeding rates by 16.2% compared to truncation selection based on EBV. However, this also resulted in short-term losses in genetic gain of 1.2% with the break-even point reached after 25 generations. In contrast, inclusion of the average kinship of an individual with individuals that would be selected based on their EBVs as an additional trait in the selection index with a weight of 17.5% resulted in no short-term losses and increased long-term genetic gain by 4.3%, while reducing inbreeding by 15.8%. Combining multiple diversity management strategies, with weights for each strategy optimized using an evolutionary algorithm, resulted in a breeding scheme with 5.1% greater genetic gain and 37.3% lower inbreeding rates than selection based on EBVs. The proposed combined strategy included the use of optimum contribution selection, weighting of SNP effects based on allele frequency, average kinship as a trait in the selection index, avoiding matings between related individuals, and lowering the proportion of selected individuals.

The combination of strategies for the management of genetic diversity in a breeding program was shown to be far superior to the use of any singular method tested in this study. As the use of strategies for management of genetic diversity and inbreeding does not necessarily lead to short-term losses in genetic gain and comes at no extra costs, it is critical for breeding companies to implement such strategies for long-term success.