310 What's Next for Genomic Selection in Dairy Cattle?. (7th December 2018)

• Record Type: Journal Article
• Authors: Lohuis, M
• Is Part Of: Journal of animal science. Volume 96(2018)Supplement 3
• Page Start: 118
• Page End: 118
• Publication Date: 2018-12-07
• Subjects: genomic selection -- evaluation; Livestock -- Periodicals; Livestock; Electronic journals; Periodicals; 636.005
• DOI: 10.1093/jas/sky404.259
• Languages: English
• ISSNs: 0021-8812

Abstract

Abstract: Genomic selection has transformed dairy genetic improvement from progeny testing to genomic testing. In Canada, genetic improvement rates have increased 2 to 3-fold due to improvements in generation interval and selection intensity. In some cases, unfavorable genetic trends in low heritability traits such as female fertility and disease tolerance have been reversed. Unfortunately, rates of inbreeding have accelerated due to rapid generational turnover. Currently, genome-tested young bull use has reached 70% and continues to rise. Due to intense preselection of bull parents the potential for bias in genomic prediction is increasing. Furthermore, as modern dairy farms withdraw from milk-recording programs and focus on in-house data collection and management tools, there is a risk that accuracy of genomic predictions could decrease over time. To compensate, more emphasis will be needed on genotyping and phenotyping unselected reference populations to maintain genomic evaluation accuracy and unbiasedness. Fortunately, traits that were previously too expensive, difficult or time-consuming to phenotype, may now be feasible in focused reference populations. AI companies may need to help support reference populations to maintain the integrity of conventional dairy traits but may also gain data on valuable novel traits such as feed efficiency, disease resistance and heat tolerance. New genomic analysis methods and technologies are helping scientists to better understand gene function. It has been shown such information can be incorporated into genomic evaluation to increase accuracy of predictions substantially and to reduce the risk of deleterious/lethal genes to be passed to the next generation. Previously, the infinitesimal model or "black box" approach was the preferred model for selection; however, genomic analysis tools may help breeders better understand the underlying genetic architecture of traits and incorporate this information into breeding programs. Additionally, opportunities exist to improve evaluation models to account for allelic effects and interactions.