Genotyping more cows increases genetic gain and reduces rate of true inbreeding in a dairy cattle breeding scheme using female reproductive technologies

Abstract

Both small dairy cattle populations and dairy cattle populations with a low level of linkage disequilibrium (LD) suffer from low reliability of genomic prediction. In this study, we investigated whether adding more genotyped cows to the reference population influences the rate of genetic gain and rate of inbreeding by affecting the reliability. A standard breeding program with a large reference population and high LD, which mimicked a breeding program for Danish Holstein population, was simulated as a reference. A Danish Jersey population with a small reference population and high LD and a Red Dairy Cattle population with a large reference population and low LD were also simulated. Two additional breeding programs were simulated for Danish Jersey and Red Dairy Cattle populations, where 2,000 additional genotyped cows were included in the population for genomic selection. All 5 simulated breeding programs were initiated by a founder population to generate LD resembling the real LD pattern, followed by a 20-yr conventional progeny-testing scheme with 1,000 or 10,000 genotyped progeny-tested bulls and a 10-yr genomic selection scheme with or without 2,000 additional genotyped cows. Evaluation criteria were annual monetary genetic gain and rate of true inbreeding. Our results showed that adding more genotyped cows to the reference in dairy cattle populations has the potential to increase genetic gain and reduce the rate of inbreeding, regardless of reference population size and level of LD. However, it is still not possible to reach the same genetic gain as in the simulated Danish Holstein population with either a small reference population or low LD. Our results also showed that in a small reference population with high LD, it is difficult to manage inbreeding because of lower accuracy compared with the simulated Danish Holstein population and a smaller number of relevant families to select from. Therefore, breeding strategies need to be chosen to match population size and structure. The rate of true inbreeding is always underestimated by pedigree inbreeding and even more in genomic breeding programs, indicating that some forms of genome-wide inbreeding, instead of pedigree-based inbreeding, should be used to monitor inbreeding when genomic selection is implemented.

Genomic selection for any dairy breeding program via optimized investment in phenotyping and genotyping

“…A larger investment into genotyping also increased update and size of the training population, which assisted in achieving genetic gain. This is in agreement with Thomasen et al [21] who showed that adding more cows yearly to the training population increases genetic gain.…”

Genomic Breeding Programs Realize Larger Benefits by Cooperation in the Presence of Genotype × Environment Interaction Than Conventional Breeding Programs

“…The founder population was simulated for 3,000 discrete generations, and the numbers of both sexes always remained the same, and it was always random mating with replacement between males and females. The detailed simulation set-up for founder population applied in this study was same as the "diverse" breeding program tailed for Red dairy cattle in Thomasen et al (2020), which intensively elaborates information such as the number and the length of chromosomes, the initial numbers and distribution of QTL and SNP markers, set-up regarding expansion and bottleneck of the population size, the rule of descendants inheriting alleles, the selection of the final QTL and SNP markers, and so on. At generation t = −1, 2,000 QTL and 40,000 SNP loci were obtained; chromosomes from 300 male and 300 female founders were gathered by sex to form 30 paternal and 30 maternal pools each with 600 chromosomes.…”

“…Genomic selection breeding programs realize larger improvement in genetic gain by cooperation than PS breeding programs (Täubert et al, 2011;Thomasen et al, 2020) because of increased selection intensity and accuracy. The increased selection intensity due to incorporating external animals as candidates (Banos and Smith, 1991;Smith and Banos, 1991;Mulder and Bijma, 2006) also applies to PS breeding programs.…”

“…This supportive proof is backed up by two underlying findings. First, GS breeding programs realized lower rate of inbreeding than conventional breeding programs either with cooperative breeding strategy or independent breeding strategy (Thomasen et al, 2020), because GS increases estimation accuracy of the Mendelian sampling term, thus allows for better differentiation within families and leads to lower coselection of sibs. More precisely, GS captures partly the Mendelian sampling term for non-phenotyped individuals and allows more balanced selection (utilizing between-and within-family variation simultaneously) than PS, which does not capture the Mendelian sampling term at all for non-phenotyped individuals (Daetwyler et al, 2007).…”

Intensified Use of Reproductive Technologies and Reduced Dimensions of Breeding Schemes Put Genetic Diversity at Risk in Dairy Cattle Breeds

“…Genetic diversity in French dairy cattle breeds is typically monitored using pedigree-based inbreeding estimates ( VARiabilité génétique des RUMinants et des Equidés , genetic variability in ruminants and equines, VARUME) [ 13 ]. However, more accurate and comprehensive evaluations can be obtained using genomic data [ 6 , 14 , 15 ]. In particular, inbreeding estimates based on genotypes and runs of homozygosity (ROH), in particular, have been shown to be as efficient, if not more, in evaluating and managing genetic diversity [ 16 ].…”