Dairy cattle

Genome-wide association study of saturated, mono and polyunsaturated milk fatty acids in Canadian dairy cows using Bayesian models

---

Abstract

In human nutrition, bovine milk fat is an essential source of fatty acids (FA), which can be classified based on the saturation of their carbon chain. Milk fat consists of around 70% of saturated FA (SFA), 25% monounsaturated FA (MUFA), and 5% polyunsaturated FA (PUFA). FA has been associated with cardiovascular disease risk and beneficial (anticancer) effects on human health; therefore, studies in dairy production systems were conducted to identify efficient strategies to optimize beneficial milk components, such as milk fatty acids. Identification of genomic regions, preferably individual genes, responsible for genetic variation in milk fat composition will enhance understanding of the biological pathways involved in fatty acid synthesis. The genome-wide association study (GWAS) is a technique for identifying genomic regions and causal genes for traits by correlating large amounts of high-density SNP genotypes with phenotypic data. This study aimed to perform a GWAS using the BayesA, BayesB, BayesC, and Bayesian Lasso models, based on 76,355 genotyping-by-sequencing SNP marker genotypes from 695 Canadian Holstein dairy cows, to identify genomic regions and markers associated with MUFA, PUFA, and SFA traits in bovine milk. The BGLR package in R was used to estimate SNP effects, defined based on the Bos taurus genome assembly (UMD 3.1). The absolute values of SNP marker effects from BayesA, BayesB, BayesC and Bayesian Lasso models accounted for by genome locations were used to create Manhattan plots for MUFA, PUFA and SFA traits in Canadian Holstein cows. The rs numbers of the top 10 ranked SNP markers were indicated on Manhattan plots for MUFA, PUFA, and SFA traits. SNP markers with the highest model frequency and marker effects were clearly separated from other markers in BayesA, BayesB, BayesC, and Bayesian Lasso models. Top 10 SNP markers having the highest marker effects with reference SNP cluster ID (rs) numbers and chromosome positions are given in Table 1, 2 and 3 for MUFA, PUFA and SFA traits, respectively. As seen in Tables 1, 2, and 3, 60% or more of the top 10 SNP markers from BayesA, BayesB, BayesC, and Bayesian Lasso models were found to be similar in the GWAS of MUFA, PUFA, and SFA traits. Also, the SNP markers of rs110638778, rs448130412 and rs133629644 on chromosome 14, rs137789897 and rs134105273 on chromosome 22 and rs42111624 and rs42111623 on chromosome 26 were determined to be common in GWAS of MUFA, PUFA and SFA traits.

Keywords: 2026

How to Cite:

Peters, S., Kızılkaya, K., Ibeagha-Awemu, E. & Zhao, X., (2026) “Genome-wide association study of saturated, mono and polyunsaturated milk fatty acids in Canadian dairy cows using Bayesian models”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2287314. doi: https://doi.org/10.31274/wcgalp.24261

Rights: 1