The impact of genomic selection on genetic diversity and genetic gain in three French dairy cattle breeds

Abstract

Background: In France, implementation of genomic evaluations in dairy cattle breeds started in 2009 and this has modified the breeding schemes drastically. In this context, the goal of our study was to understand the impact of genomic selection on the genetic diversity of bulls from three French dairy cattle breeds born between 2005 and 2015 (Montbéliarde, Normande and Holstein) and the factors that are involved.

Methods: We compared annual genetic gains, inbreeding rates based on runs of homozygosity (ROH) and pedigree data, and mean ROH length within breeds, before and after the implementation of genomic selection.

Results: Genomic selection induced an increase in mean annual genetic gains of 50, 71 and 33% for Montbéliarde, Normande and Holstein bulls, respectively, and in parallel, the generation intervals were reduced by a factor of 1.7, 1.9 and 2, respectively. We found no significant change in inbreeding rate for the two national breeds, Montbéliarde and Normande, and a significant increase in inbreeding rate for the Holstein international breed, which is now as high as 0.55% per year based on ROH and 0.49% per year based on pedigree data (equivalent to a rate of 1.36 and 1.39% per generation, respectively). The mean ROH length was longer for bulls from the Holstein breed than for those from the other two breeds.

Conclusions: With the implementation of genomic selection, the annual genetic gain increased for bulls from the three major French dairy cattle breeds. At the same time, the annual loss of genetic diversity increased for Holstein bulls, possibly because of the massive use of a few elite bulls in this breed, but not for Montbéliarde and Normande bulls. The increase in mean ROH length in Holstein may reflect the occurrence of recent inbreeding. New strategies in breeding schemes, such as female donor stations and embryo transfer, and recent implementation of genomic evaluations in small regional breeds should be studied carefully in order to ensure the sustainability of breeding schemes in the future.

Fig. 1

Breeding schemes of bulls under progeny testing (a) and genomic (b) selection

Fig. 2

Trends of the different parameters estimating genetic gain and genetic diversity. The yellow line represents the mean for each birth year and the error bars the standard deviation. Total merit indices (ISU) (a), inbreeding (b) and kinship (c) based on pedigree data, inbreeding based on runs of homozygosity (ROH) (d) and mean ROH length (in kb) (e) were retrieved or computed for progeny-tested bulls born between 2005 and 2010 and for marketed bulls born between 2012 and 2015, for Montbéliarde, Normande and Holstein

Fig. 3

Pedigree-based inbreeding calculated from the last five generations. The yellow line represents the mean pedigree-based inbreeding calculated from the last five generations for each birth year and the error bars the standard deviation. Pedigree-based inbreeding calculated from the last five generations was computed for progeny-tested bulls born between 2005 and 2010 and for marketed bulls born between 2012 and 2015, for Montbéliarde (a), Normande (b) and Holstein (c)

Fig. 4

Parental generation intervals I₁ (in months). I₁ were defined as the difference between the birthdate of an individual and its parents’ (in months). The yellow line represents the mean for each birth year and the error bars the standard deviation. Generation intervals I₁ were computed for progeny-tested bulls born between 2005 and 2010 and for marketed bulls born between 2012 and 2015, for Montbéliarde (a), Normande (b) and Holstein (c)

Fig. 5

Proportion of bulls born each year in comparison with 2005. Proportions were computed for progeny-tested bulls born between 2005 and 2010 and for marketed bulls born between 2012 and 2015, for Montbéliarde (green), Normande (blue) and Holstein (red)

Fig. 6

Effective number of bulls of three French dairy cattle breeds. Effective number of bulls were computed for progeny-tested bulls born between 2005 and 2010 for progeny testing selection (blue) and for marketed bulls born between 2012 and 2015 for genomic selection (red) for Montbéliarde, Normande and Holstein. Black bars represent 95% confidence intervals, generated by random resampling with replacement of the datasets (bootstrap with 1000 iterations)

Fig. 7

Effective/census number of bulls ratios of three French dairy cattle breeds. Effective number of bulls/Number of bulls ratios $N{e}_{\text{breed, selection}}/n_{\text{breed, selection}}$ were computed for progeny-tested bulls for individuals born between 2005 and 2010 for progeny testing selection (blue) and for marketed bulls born between 2012 and 2015 for genomic selection (red) for Montbéliarde, Normande and Holstein. Black bars represent 95% confidence intervals, generated by random resampling with replacement of the datasets (bootstrap with 1000 iterations)