Comparison of selective genotyping strategies in genomic selection programs for broilers using stochastic simulation

Abstract

Breeding programs using genomic information have become common in broilers. In broilers, the main traits for selection are body weight and feed efficiency. These traits are measured in both sexes and before sexual maturity. Thus, increases in genetic gain in broiler breeding programs due to the use of genomic information are attributable primarily to the increased accuracy of predicted breeding values of potential parents. As not all birds can be genotyped due to economic constraints, an optimal genotyping strategy needs to be chosen. The objective of this study was to investigate the rates of genetic gain and inbreeding obtained by varying three factors: the percentage of birds genotyped (PG), the percentage of birds genotyped randomly (PRG) or selectively, and the percentage of males among genotyped birds (PMG). Stochastic computer simulation with a full factorial experimental design was used. Significant interaction among the factors (PG, PRG, and PMG) was observed for the rate of genetic gain in body weight and aggregate breeding value, but not for the gain in residual feed intake or the rate of inbreeding per generation. Our findings indicate that the PG, PRG, and PMG need to be considered when selecting a genotyping strategy for a broiler breeding program. If available resources allow only to genotype a small percentage (e.g., 2.5% or 5%) of all birds, the genotyping of 50–100% of male birds and selection of birds to be genotyped using phenotypic information is best. If resources allow to genotype more (e.g., ≥20%) candidates, genotyping of equal numbers of each sex, and low PRG level (i.e., preselection of the majority of candidate based on performance) is best. Provided that a proper genotyping strategy is chosen, we conclude that the incorporation of genomic information in broiler breeding programs can substantially increase the rate of genetic gain.