Changes in the genome architecture of two groups of dairy bulls with marked differences in their direct genomic breeding values for production traits in the UK

Abstract

Genomic selection has resulted in a rapid rate of genetic progress in dairy cattle in the last decade which could partly be attributed to a marked reduction in generation interval. For instance, in the UK the average age of a bull when their 100^th daughter was born has decreased from over 6 years prior to the introduction of genomics to under 4 years in 2022. It is however not clearly understood if this rapid rate of genetic progress has also been accompanied by changes in the genome architecture in terms of degrees of heterozygosity, allele frequencies and linkage disequilibrium (LD) structure. This study examines changes in these parameters in 9,202 bulls born between 2009 and 2014 in the reference population for production traits and 94,204 young bulls with no daughters born between 2019 and 2023. The mean difference in direct genomic breeding values (DGV) between these reference and young bulls (YG) were 357.7, 24.5 and 16.1 kg for milk, fat and protein yields respectively. The SNP panel used in the UK evaluations consists of 79,051 SNPs and the proportion of SNPs with 0 to 1, 2 to 5, 6 to 10, 11 to 15 and greater than 15 percent change in their allele frequencies between both the reference and young bull groups were 15, 32, 29, 15 and 9% respectively. The SNPs with at least 15% change in allele frequencies accounted for 34, 43 and 39 percent of the mean difference in DGV between reference and young bulls for kg milk, fat and protein respectively, while the corresponding values for SNPs with 0 to 1 % change in allele frequencies were less than one percent for all three traits. In absolute terms the correlation between differences in mean DGV between reference and young bulls and changes in allele frequencies at the chromosome level was about 0.65 for the three traits. Thus, the rapid rate of genetic progress due to genomic selection is significantly changing allele frequencies. The rate of LD decay was similar for both groups of bulls, but tended to be higher for YG, suggesting stronger selective pressures and/or lower effective population size. The increased rate of inbreeding in shorter ROH but slower increase in longer ROH in YG seem to imply that recent inbreeding is being better controlled than very ancient inbreeding.