Implementation of advanced Optimum Contribution Selection in small-scale breeding schemes: prospects and challenges in Vorderwald cattle

Abstract

Vorderwald cattle are a regional cattle breed from the Black Forest in south western Germany. In recent decades, commercial breeds have been introgressed to upgrade the breed in performance traits. On one hand, native genetic diversity of the breed should be conserved. On the other hand, moderate rates of genetic gain are needed to satisfy breeders to keep the breed. These goals are antagonistic, since the native proportion of the gene pool is negatively correlated to performance traits and the carriers of introgressed alleles are less related to the population. Thus, a standard Optimum Contribution Selection (OCS) approach would lead to reinforced selection on migrant contributions (MC). Our objective was the development of strategies for practical implementation of an OCS approach to manage the MC and native genetic diversity of regional breeds. Additionally, we examined the organisational efforts and the financial impacts on the breeding scheme of Vorderwald cattle. We chose the advanced Optimum Contribution Selection (aOCS) to manage the breed in stochastic simulations based on real pedigree data. In addition to standard OCS approaches, aOCS facilitates the management of the MC and the rate of inbreeding at native alleles. We examined two aOCS strategies. Both strategies maximised genetic gain, while strategy (I) conserved the MC in the breeding population and strategy (II) reduced the MC at a predefined annual rate. These two approaches were combined with one of three flows of replacement of sires (FoR strategies). Additionally, we compared breeding costs to clarify about the financial impact of implementing aOCS in a young sire breeding scheme. According to our results, conserving the MC in the population led to significantly (P < 0.01) higher genetic gain (1.16 ± 0.13 points/year) than reducing the MC (0.88 ± 0.10 points/year). In simulation scenarios that conserved the MC, the final value of MC was 57.6% ± 0.004, while being constraint to 58.2%. However, reducing the MC is only partially feasible based on pedigree data. Additionally, this study proves that the classical rate of inbreeding can be managed by constraining only the rate of inbreeding at native alleles within the aOCS approach. The financial comparison of the different breeding schemes proved the feasibility of implementing aOCS in Vorderwald cattle. Implementing the modelled breeding scheme would reduce costs by 1.1% compared with the actual scheme. Reduced costs were underpinned by additional genetic gain in superior simulation scenarios compared to expected genetic gain in reality (+4.85%).

Keywords: breeding costs; migrant contribution; native contribution; native kinship; regional breed.

Figure 1

Development of genetic gain of birth cohorts in different simulation scenarios and extrapolated reality for Vorderwald cattle – Development of mean estimated breeding values for the total merit index of birth cohorts in simulation scenarios and reality. Development in the real population was calculated based on the real data between 2005 and 2015. We assumed that the genetic gain will evolve linearly with an annual rate of 1.18 points/year. In simulation scenarios, we examined varying flows of replacement of sires with 10, 20 or 30 on an annual basis (FoR10, FoR20 and FoR30, respectively) in combination with two different aOCS strategies. The first one conserved the mean MC in subsequent birth cohorts (conserve-MC). The second one reduced MC with an annual rate of –0.35% (reduce-MC). Results of simulation scenarios were averaged over five replicates (±SD). aOCS = advanced Optimum Contribution Selection; MC = migrant contribution.

Figure 2

Development of MCs of birth cohorts in different simulation scenarios for Vorderwald cattle – Development of mean MC of birth cohorts in simulation scenarios. We examined varying flows of replacement of sires with 10, 20 or 30 on an annual basis (FoR10, FoR20 and FoR30, respectively) in combination with two different aOCS strategies. The first one conserved the mean MC in subsequent birth cohorts (conserve-MC). The second one reduced MC with an annual rate of –0.35% (reduce-MC). Results of simulation scenarios were averaged over five replicates (±SD). MC = migrant contribution; aOCS = advanced Optimum Contribution Selection.

Figure 3

Development of average classical and native kinship coefficients of evolving populations in different simulation scenarios for Vorderwald cattle – We examined varying flows of replacement of sires with 10, 20 or 30 on an annual basis (FoR10, FoR20 and FoR30, respectively) in combination with two different aOCS strategies. The first one conserved the mean MC in subsequent birth cohorts (conserve-MC). The second one reduced the MC with an annual rate of –0.35% (reduce-MC). Both aOCS strategies restricted the average kinship at native alleles (natKin) of the population by an upper bound (black and solid), meanwhile the classical kinship (classKin) was not managed. The graphs are subdivided for different simulation scenarios (colour) and both kinship coefficients (solid and dashed). Results of simulation scenarios were averaged over five replicates (±SD). aOCS = advanced Optimum Contribution Selection; MC = migrant contribution.

Figure 4

Bar graph of annually contributing sires in different simulation scenarios for Vorderwald cattle – We examined varying flows of replacement of sires with 10, 20 or 30 on an annual basis (FoR10, FoR20 and FoR30, respectively) in combination with two different aOCS strategies. The first one conserved the mean MC in subsequent birth cohorts (conserve-MC). The second one reduced MC with an annual rate of –0.35% (reduce-MC). The bar graphs are subdivided for FoR strategies. Bars visualise the average of annually contributing sires averaged over five replicates (±SD). aOCS = advanced Optimum Contribution Selection; MC = migrant contribution.