Genomic architecture of non-nominal offspring risk in a simulated composite cattle breed

BACKGROUND: Composite cattle breeds are designed to combine complementary attributes from different founder breeds while retaining part of the heterosis generated by crossbreeding. Brangus cattle are nominally defined as 5/8 Angus and 3/8 Brahman, but the same global breed composition can be produced by different genomic architectures. This distinction may be important because animals with the same Brahman ancestry proportion may differ in ancestry heterozygosity, homozygous ancestry composition and segregation behaviour when used as parents. The objective of this study was to use controlled simulations to evaluate whether genomic architecture affects the risk of producing offspring outside the nominal Brangus ancestry range. RESULTS: We simulated two alternative pathways to the same expected Brahman ancestry proportion of 0.375. Although both pathways produced similar mean Brahman ancestry, the direct pathway generated a broader ancestry distribution than the indirect pathway, with a higher proportion of individuals outside the nominal interval. After filtering candidate sires to the nominal interval, formation-pathway-dependent genomic architectures remained evident. In a mixed inter se mating analysis extended to 100 generations and three population-size scenarios, pathway-specific ancestry-state differences were progressively eroded and the standard deviation of Brahman ancestry proportion declined after an early maximum, but non-nominal frequencies remained non-negligible over the simulated time horizon. Family-level simulations showed that matings between nominal parents differed substantially in their probability of producing non-nominal daughters. This risk depended on both the distance between the expected family mean and the target ancestry and the block-weighted segregation potential of the parental genomes. An exploratory quasibinomial model supported positive associations between non-nominal risk and displacement from the target and high-variance parental block structures. Finally, a heuristic-assisted simulation-informed sire allocation strategy increased the proportion of nominal daughters relative to random sire allocation, with the greatest gains obtained from broader sire catalogues. CONCLUSIONS: Nominal breed composition does not define a unique genomic state. In simulated Brangus cattle, animals that were equivalent in global ancestry could differ in genomic architecture and in the offspring ancestry distributions generated by specific matings. These results suggest that composite-breed management should consider not only the ancestry proportion of individual animals, but also the expected outcome of sire-dam pairs.