Genomic evaluation of bulls for daughters’ milk traits in Russian black-and-white and Holstein cattle population through the validation procedure

The rapid development of molecular genetic methods in animal breeding over the past ten years has given rise to an increase of the selection intensity at the population level. Expansion of the economically useful traits spectrum of dairy cattle allowed increasing the opportunities for breeding to improve the cows’ health and for studying the essence of the metabolic synthesis of milk components. The purpose of this study was to verify the effectiveness of genomic forecasting in the development of the concept of dairy cattle genetic assessment in the regional and national aspects. The study for bulls’ estimations in Russian Black-and-White improved by Holstein and Holstein breeds by simulation of breeding process using 124 herds of the Moscow and Leningrad regions was carried out. The effectiveness of genomic prediction as compared to the parent averages (PA) and the estimated breeding values (EBV) of sires have been shown. The selection of testing bulls based on the genomic information corrects PA and is to refine EBV that is further obtained by using progeny. Repeatability of genomic EBV was obtained through validation of parentage and genomic information for 100 sires with data for at least 300 daughters. This dataset lay down the core of the newly created Russian regional reference group of dairy cattle. For calculating the additive relationship matrix 1050 ancestors was used. For genomic relationship matrix, 39818 nucleotide polymorphisms were taken into analysis. Based on the REML, BLUP SM, GBLUP methods the procedures to assess of animals were carried out. The average annual genetic trend for milk production traits in the studied populations from 1987 to 2006 was +60 kg, +2.5 kg, +1.5 kg by milk yield, milk fat and milk protein, respectively. It was found that the repeatability of genomic estimates was ranging from 0.371 to 0.606 for milk production traits, which on average exceeded the PA value by 0.147. The accuracy of the evaluation obtained by progeny tested bulls ranged from 0.879 to 0.900 that was higher than the genomic prediction by 0.405 units. The principles of creating the reference population based on the analysis of multi dimension scaling and genetic distances were studied. The distinguish between two regional populations (Moscow and Leningrad regions) was Fst = 0.0025. The decay of the linkage disequilibrium between the markers at distances up to 1000 kb is shown. In distance from 5 up to 70 kb the linkage level was get the maximum values from 0.20 to 0.54. In the framework of the metabolic pathways study for the milk components synthesis genetic parameters and mean least square estimates were obtained for the extended milk composition: lactose (h2 = 0.18), dry matter (h2 = 0.10), solids-not-fat (h2 = 0.19), milk freezing point (h2 = 0.06), somatic cells score (h2 = 0.10) and milk urea (h2 = 0.04). The values of additive genetic variances have been get indicating the objective possibilities of using them in Russian dairy cattle breeding sector. To obtain reliable whole-genome associations a further replenishment of the database of the cows’ milk component will be carried out using additional spectra. The complex studies have grounded approaches to the use of genomic estimations, the principles of reference population extension and widening list of features for the quantitative and qualitative milk composition assessment.