Modeling of milk yield lactation curves, milk components, and metabolic metabolites of Holstein cows

In recent years, there have been many population-based studies devoted to the analysis of the component composition of cow’s milk, aimed at a more detailed study of the variability of indicators under the influence of genetic and environmental factors. The research base consisted of observations from 14 breeding herds of Holstein cattle. Population-genetic studies of the components of cow’s milk were carried out during the control milking periods (the total number of milk samples is more than 36 thousand pieces). 11.5 thousand heads of animals were examined with indicators of biomarkers such as acetone, BGB and urea. Based on the use of a nonlinear function, the lactation curves of daily milk yield, the main components of milk and metabolites of metabolism were modeled. The standard form of the lactation curve description showed the highest level of accuracy — 93.5%, while the percentage of fat and protein describe the opposite forms. The coefficient of determination of the content of the mass fraction of protein during lactation was 83.9%, and fat — only 63.7%. At the same time, the percentage of reliability of the lactation curve model for the content of the mass fraction of lactose was 68.4%. The content of traces of molar urea concentration in milk is modeled somewhat worse, and the coefficient of determination is 46.8%. The reliability of the model of standard lactation curves for traces of acetone and BGB in the milk of Holstein cows was 41.6% and 26.0%, respectively. The results of the analysis of the dynamics of milk yield changes and the application of the developed regression equations of lactation curves demonstrated the prospects of their use for low-heritable traits in order to determine the functional qualities of animals in the context of the course of lactation by period.