Animal Model for Genetic Evaluation of Dairy Cattle in Japan

An animal model for evaluating the genetic merits of dairy cattle in production traits (milk, fat, SNF, and protein yield and percent) and type traits (5 score traits and 15 linear traits) using large field data, such as records of the Dairy Herd Improvement Program (DRIP), was developed. The model for evaluating production traits included the effects of management group (herd-year-country where sire keptparity), calving month by area, age at calving, breeding value, random permanent, and random residual. The model for evaluating type traits included the effects of judgment group (herd-judge-date), age at judgment, and stage of lactation at judgment. The breeding value is a random additive genetic effect plus contributions from genetic groups formed by unknown and unidentified parents. The animal model was applied to actual records on trial. The numbers of records were 2, 335, 401 for 912, 310 cows in production traits, and 120, 287 in type traits. The mixed model equation was solved iteratively by the indirect approach. The correlation between the sires' breeding values from the trial and the official proofs under the MGS model was high in every traits (_??_0.89). Genetic and management trends in the production traits nation wide were eatimated. The annual improvement in cows' breeding value and management efficiency in milk were 46.6kg, and 109.6kg (for first lactation) and 147.7kg (for later lactation), respectively, over the last 10 years.