Genomic selection improves the possibility of applying multiple breeding programs in different environments

摘要

One joint breeding program (BP) for different dairycattle environments can be advantageous for geneticgain depending on the genetic correlation between environments(r_g). The break-even correlation (rb) refersto the specific r_g where genetic gain with 1 joint BPis equal to the genetic gain of 2 environment-specificBP. One joint BP has the highest genetic gain if r_g ishigher than rb, whereas 2 environment-specific BP havehigher genetic gain if r_g is lower than rb. Genetic gainin this context is evaluated from a breeding company’sperspective that aims to improve genetic gain in bothenvironments. With the implementation of genomicselection, 2 types of collaboration can be identified:exchanging breeding animals and exchanging genomicinformation. The aim of this study was to study geneticgain in multiple environments with different breedingstrategies with genomic selection. The specific aimswere (1) to find rb when applying genomic selection; (2)to assess how much genetic gain is lost when applyinga suboptimal breeding strategy; (3) to study the effectof the reliability of direct genomic values, number ofgenotyped animals, and environments of different sizeon rb and genetic gain; and (4) to find rb from eachenvironment’s point of view. Three breeding strategieswere simulated: 1 joint BP for both environments, 2environment-specific BP with selection of bulls acrossenvironments, and 2 environment-specific BP with selectionof bulls within environments. The rb was 0.65and not different from rb with progeny-testing breedingprograms when compared at the same selection intensity.The maximum loss in genetic gain in a suboptimalbreeding strategy was 24%. A higher direct genomicvalue reliability and an increased number of genotypedselection candidates increased genetic gain, and theeffect on rb was not large. A different size in 2 environmentsdecreased rb by, at most, 0.10 points. Froma large environment’s point of view, 1 joint BP wasthe optimal breeding strategy in most scenarios. Froma small environment’s point of view, 1 joint BP wasonly the optimal breeding strategy at high r_g. Whenthe exchange of breeding animals between environmentswas restricted, genetic gain could still increasein each environment. This was due to the exchangeof genomic information between environments, evenwhen r_g between environments were as low as 0.4. Thus,genomic selection improves the possibility of applyingenvironment-specific BP.