Developing a genetic evaluation system for milk traits in Russian black and white dairy cattle

Mixed linear models have been applied for predicting breeding values of dairy cattle in most of the developed countries since the 1980s. However, the Russian Federation is still using the old contemporary comparison method. The objective of our study was to develop a best linear unbiased prediction (BLUP) for an animal model of breeding values for the Leningrad region. We tested both a first-lactation model (FLM) and a multi-lactation repeatability model (MLM). The data included milk records of 206 114 cows from 49 herds. Estimated heritabilities from FLM were 0.24, 0.20, and 0.20 for milk, protein, and fat yields, respectively, and 0.18, 0.19, and 0.20 from MLM. Repeatabilities were 0.34 for milk yield and 0.31 for both fat and protein yields. Genetic trends were similar for both models (FLM vs MLM): 59 vs 56 kg year-1 for milk, 1.90 vs 1.84 kg year-1 for fat, and 1.67 vs 1.62 kg year-1 for protein yield during 2000–2016. Based on the difference between the genetic trends in FLM and MLM, the applied BLUP method passed the validation method I by Interbull.