Structural equation modeling for unraveling the multivariate genomic architecture of milk proteins in dairy cattle

The aims of this study were to investigate potential functional relationships among milk protein fractions in dairy cattle and carry out a structural equation model (SEM) GWAS provide decomposition total SNP effects into direct mediated by traits that are upstream phenotypic network. To achieve these aims, we first fitted mixed Bayesian multitrait genomic infer the correlations 6 nitrogen [4 caseins (CN), namely κ-, β-, αS1-, αS2-CN, 2 whey proteins, β-lactoglobulin (β-LG) α-lactalbumin (α-LA)], population 989 Italian Brown Swiss cows. Animals genotyped with Illumina BovineSNP50 Bead Chip v.2 (Illumina Inc.). A network approach using max-min hill-climbing (MMHC) algorithm was implemented dependencies or independence traits. Strong negative found between β-CN αS1-CN (−0.706) κ-CN (−0.735). application MMHC revealed seemed directly indirectly influence all other fractions. By integrating SEM-GWAS, identified 127 significant for κ-CN, 89 β-CN, 30 αS1-CN, 14 αS2-CN (mostly shared CN located on Bos taurus autosome 6) 15 β-LG 11), whereas no passed significance threshold α-LA. For SNP, assessed quantified contribution indirect paths marker effect. Pathway analyses confirmed common regulatory mechanisms (e.g., energy metabolism hormonal neural signals) involved control synthesis metabolism. information acquired might be leveraged setting up optimal management selection strategies aimed at improving quality technological characteristics cattle.