Integration of dominance effects into genomic models for enhancing the understanding of heterosis in dairy cattle

Molecular heterozygosity and heterosis are gaining importance in the evaluation of various species. However, the exploration of estimated heterosis based on the genome in purebred populations remains limited. We aimed at investigating the use of heterozygosity as a potential indicator of genomic heterosis in purebred populations. Using a simulated dataset for milk production and quality traits, we considered three scenarios (h² = 0.10, 0.30, and 0.50). We adapted the Genome-Wide Association Study (GWAS) to capture non-additive variations for milk yield (MY), fat percentage (FP), protein percentage (PP), casein percentage (CP), and polyunsaturated fatty acid percentage (PUFA). Heterozygosity ranged from 35.4% to 35.6% in the simulated scenario, with genomic heterosis ranging from 3.2% to 17.0%. Regression coefficients emphasized the significance of heterozygosity for genomic heterosis, varying from 4.8 to 6.07. In real data, most identified genomic regions showed consistency between additive and non-additive models. An increase of 1.96 kg/day in MY was associated with a one-unit increase in heterozygosity, along with a 0.0059% increase in PUFA. Eight genes, including the SOX5 gene associated with PUFAs, have been identified in the literature as important for human health. Selection based on heterozygosity is proposed to favor genomic heterosis, and considering dominance effects in GWAS contributes to marker and QTL identification with potential heterozygous advantages. Our study significantly contributes to understanding how heterosis can be utilized for animal selection in pure breeds using genomic information. Our findings suggest predictive approaches tailored to include genetic dominance effects in genetic evaluation and genome-wide association studies.