J. B. Cole, M. E. Tooker, P. M. VanRaden, and J. H. Megonigal, Jr.
Animal Improvement Programs Laboratory ,
Agricultural Research Service,
USDA, Beltsville, MD 20705-2350
Crossbred animals in the US dairy cattle population do not currently receive genetic evaluations. A system for storing breed composition data was developed and the software implemented. Crossbreds are animals whose sire and dam breeds are known but different. Two ways that crossbred animals may be handled in an evaluation are: they may be included in the breed-of-sire evaluation, or all animals may be combined into a single evaluation. For each crossbred cow (n=431,432) the database stores animal identification, breed of animal registry, breed fractions for 18 dairy breeds and an "unknown" category, a four-generation breed stack, a sire breed stack, and a heterosis value. Standard two-letter breed codes are used in the breed stacks. Breed fractions were calculated from pedigree information in the USDA database as far back as data were available. As a result, the crossbred table includes >70,000 crosses that occurred >4 generations back in pedigrees, resulting in one breed with a very large contribution (e.g. 96.9% HO) to the breed composition and the other with a very small contribution (e.g. 3.1% JE). To avoid confusion the data may be reformatted for presentation (e.g. 97HO3JE), which would involve a small loss of precision. When purebred sires are used, the sire breed stack completely describes the mating system and may be used to identify rotational breeding programs. If crossbred sires are used, a four-generation breed stack is needed to identify reciprocal crosses that cannot be denoted by the breed composition alone. Consider a purebred Holstein sire mated to a purebred Jersey cow. Breed composition of a daughter would be 50HO50JE with a sire breed stack of HOJE and heterosis of 1.0. If that daughter were backcrossed to a Jersey bull, the resulting calf would have a breed composition of 75JE25HO, a sire breed stack of JEHOJE, and heterosis of 0.50. This system is necessary to facilitate implementation of across-breed genetic evaluations.