Estimation & Prediction

DNA pooling and group birth weight phenotypes for genetic prediction

---

Abstract

Genomic studies in livestock often rely on individually genotyped animals, although some work has been conducted with pooled genotypes, each having advantages and limitations. Individual genotyping provides accurate estimates of allele frequencies and facilitates high-resolution genetic analyses, but it is costly when applied to large populations. In contrast, pooled genotyping offers a cost-effective means to capture population-level variation, though at the expense of individual-level information. In this study, we combined genotypes derived from both individual cattle and pooled DNA samples into a single analysis. We integrated individual genotypes (n=6,460) and 16 pooled genotypes from the U.S. Meat Animal Research Center (USMARC) Selection for Functional Alleles cattle population. Individual genotypes of the 954 pooled animals were also available. A univariate animal model fitting a genomic relationship matrices (GRM) was used for the trait of birth weight (BWT) to jointly estimate genomic relationships and breeding values for individuals and pools. Model performance was evaluated by comparing results from the combined dataset with analyses conducted on individual and pooled data separately. The combined approach yielded similar estimates of heritability, with an estimate of 0.499 using individual animals and 0.496 with the combined GRM that included both pools and individuals. Using only the individual genotypes of the pooled animals yielded a heritability estimate of 0.201. The Spearman rank correlation between the estimated breeding values using pooled genotypes and the average of the individual pooled animal genotypes was 0.889. The Spearman rank correlation between the estimated breeding values of individual genotypes and the combined dataset animal genotypes was 0.857. This integrative framework provides a practical and scalable approach for livestock genomics and breeding programs, enabling researchers to maximize the utility of available genotypic resources. Future work will focus on refining statistical methods to better account for uncertainty in pooled allele frequency estimates and on exploring the applicability of this approach in a bi-variate framework. The USDA is an equal opportunity provider and employer.

Keywords: 2026

How to Cite:

Keele, J., Kuehn, L., Russell, C., Snelling, W. & Spangler, M., (2026) “DNA pooling and group birth weight phenotypes for genetic prediction”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2286876. doi: https://doi.org/10.31274/wcgalp.24152

Rights: 1