234 Award Talk: From genomics to herd improvement: Pioneering strategies in dairy cattle breeding.

Abstract The widespread adoption of genomic selection by the dairy industry has been an unparalleled success, as shown by a $4 billion return on an initial investment by US taxpayers of $100 million in its first decade. While the most immediate and visible benefits of genomic tools are their effects on the accuracy and intensity of genetic selection – which provide cumulative and permanent benefits to dairy cows, producers, and consumers – those are only the tip of the iceberg. These tools can be used to preserve genetic diversity and managing harmful defects, enable effective selection for new phenotypes with high recording costs, permit the validation of key statistical and biological assumptions, support discovery science that was not previously possible in large ruminants, and act as a foundation for sophisticated farm management strategies. The impacts of genomic selection on breeds commonly used with intensive dairying systems are well documented and, despite the rapid decrease in generation intervals and increase in rates of genetic gain, we are far from estimated selection limits. It is likely that management will be limiting before genetics are. Coefficients of genomic inbreeding and runs of homozygosity provide high-resolution snapshots of individual genetic diversity, which is still a challenge to manage at the population level. Haplotype information, performance data from carrier-to-carrier matings, and low-cost functional validation of early embryonic lethals can be used together to identify and manage genetic defects. The ability of genotyping to provide acceptable genomic reliabilities for selection using a modest number of phenotypes has enabled the rapid adoption of traits that are expensive to measure, such as feed intake and greenhouse gas emissions, into selection programs. At this point, the absence of market values, e.g., for carbon credits, is more limiting in some populations than the number of animals with performance records. Many studies over the past decade have confirmed that the infinitesimal alleles, while inaccurate in some respects, remains a useful model of biology and few true quantitative trait loci have been discovered. Large-scale community projects, such as 1000 Bull Genomes, FAANG, and FarmGTEx, have demonstrated the value of coordinated effort for functional biology.