PSIII-16 - Application and economic analysis of pooled genome-wide association studies in commercial feedlot cattle

Abstract: Genotyping an individual animal provides a specified identity of their genetic makeup, which when used in a commercial beef operation, may allow for precise predictions centered around trait quality and production. However, conventional genotyping occurs at the individual level and is fiscally and logistically taxing for commercial cattle producers for whole-herd testing. Therefore, our research objective was to utilize DNA pooling methodology and genome-wide association studies (GWAS) to introduce economically feasible genomic testing in commercial feedlot cattle.

A total of 1,956 commercial mixed-breed feedlot cattle (1,715 steers; 241 heifers) with accompanying complete carcass date were enrolled in the project after having their DNA extracted from tissue sampling units collected during feedlot enrollment. Isolated DNA was stratified into 79 pools (mean pool size = 25 individuals) for genotyping via GGP Bovine 100K array (Illumina); pooling was designed to replicate a prospective timeline, utilizing lot number, arrival date, and feedlot entry weight to stratify DNA pools. Data preparation with outlier filtering (OF) and repeated measures (RM) reduced phenotypic variance within pools by removing pools containing outliers, defined as values greater than 150% of the IQR outside of the IQR, and replicating a pool genotype by the number of individuals in a pool to compare to individual phenotype, relying on the concept that pooled DNA is a proxy average of individual genetic makeup, respectively. To complete the GWAS, phenotype files were built using pool average and individual calculated yield grade (CYG) and marbling (MARB) values. To evaluate economic feasibility, we performed a cost comparison of the pooling process itself as compared to individual genotyping, in addition to a cost-benefit analysis for prospective pooling to display the potential increase in income because of the investment in genotyping.

Completed GWAS identified 20, 10, 27, and 16 significant SNPs associated with RM CYG, OF CYG, RM MARB, and OF MARB, respectively (p ≤ 0.001). The estimated total cost of genotyping the pooled DNA in the project ($23,460) was 4.5 times less than the estimated cost of genotyping the same cattle individually ($105,354), and the mean potential benefit-cost ratio of utilizing pooled genotypes to make management decisions was 2.33 (SD = 1.91).

Our research demonstrated the successful use of DNA pooling in a commercial feedlot setting and showcased a need for further research and more exploration to better define a successful prospective pooling process. Although the dataset was highly variable, these biologically logical associations signify relative success of utilizing pooled DNA in GWAS. The economic efficiency of the DNA pooling process itself warrants further testing, as the economic benefits the process can provide producers outweigh costs.