Skip to main content
ARS Home » Plains Area » Clay Center, Nebraska » U.S. Meat Animal Research Center » Genetics and Animal Breeding » Research » Publications at this Location » Publication #370276
Research Project: Developing a Systems Biology Approach to Enhance Efficiency and Sustainability of Beef and Lamb Production

Location: Genetics and Animal Breeding

Title: Development of a low-pass sequencing platform to support genomic selection in livestock

Author
item PICKRELL, JOSEPH - GENCOVE, INC.
item LI, JEREMIAH - GENCOVE, INC.
item HOFF, JESSE - GENCOVE, INC.
item Kuehn, Larry
item Snelling, Warren

Submitted to: Annual International Plant & Animal Genome Conference
Publication Type: Abstract Only
Publication Acceptance Date: 10/30/2019
Publication Date: 1/2/2020
Citation: Pickrell, J., Li, J.H., Hoff, J.L., Kuehn, L.A., Snelling, W.M. 2020. Development of a low-pass sequencing platform to support genomic selection in livestock [abstract]. International Plant & Animal Genome XXVIII Conference, January 11-15, 2020, San Diego, California. Abstract Number W218. Available at: https://plan.core-apps.com/pag_2020/abstracts

Interpretive Summary:

Technical Abstract: Genotyping platforms for breeding applications are required to be cost-effective and high throughput; the general approach to meeting these criteria involves identifying a small subset of variant sites to be measured on a genotyping array. Low-pass whole genome sequencing combined with genotype imputation can, in principle, allow for the measurement of the entire genome in a cost-effective manner. In order to build a platform for low-pass sequencing in cattle, we constructed a cattle haplotype reference panel by re-analysis of 946 genome sequences from 14 cattle breeds. All sequences were jointly genotyped using GATK 4, producing a total of around 60M markers after filtering. The identified sites contain >95% of the sites on the BovineSNP50 and BovineHD assays. We evaluated the accuracy of genotype imputation from low-pass sequencing data using this reference panel by downsampling sequencing data to coverage levels ranging from 4x to 0.4x. At a sequencing depth of 0.4x, concordance between imputed low-pass sequencing data and directly genotyped sites was over 99% in most B. taurus breeds and 97% in Brahman. Finally, in a sample of 100 cattle we compared the genomic relationship matrix and genomic predictions generated using imputed low-pass sequencing data and the SNP50 array, finding them to be effectively equivalent. We describe future prospects for genomic prediction using low-pass sequencing, including greater compatibility as a method across populations, and flexible integration with historic and future markers sets. We suggest that re-analysis of sequencing data in breeding animals will allow for continuous updates of the marker training set used to optimize genomic prediction accuracy, on a population- and trait-specific basis.