Estimation of genetic parameters of lactation curves of Iranian Holstein dairy cattle

Introduction: The graphical representation of milk production in a lactation period in the form of a diagram is called a lactation curve. The lactation curve shows the biological efficiency of an animal and can be a tool for management and selection. The shape of the lactation curve is characterized by the initial incremental slope of the curve, the peak production value, the peak time, the slope of the curve after the peak production (continuation of lactation) and the length of the lactation period. The shape of the lactation curve provides valuable information that is essential for evaluating the biological and economic performance of an individual or herd and is useful for genetic evaluation and screening, health testing, nutritional management decisions, and planning purposes. Estimating the genetic parameters of the lactation curve and knowing the values of the parameters can be useful in designing breeding programs and predicting genetic improvements and herd management. According to the very limited research related to the use of the MilkBot function in fitting the lactation curves and the genetic investigation of curve parameters for Iranian Holstein cows, this research aims to fit the lactation curve of the first, second and third abdominal cows and the genetic investigation of the parameters of the lactation curve of cows was designed and executed.
Material and Methods: In this study, 1650669, 1198923 and 781859 test day records of 122455, 91064 and 60024 first, second and third calving cows belonging to 17 herds between 2012 and 2023 that were collected by Arin Delta Gen International Company were used. The data were edited based on the following manner: lactation days between 5 and 325 days, the number of test day records for each animal at least 5 records, calving age for cows in the first parity between 21 and 48 months, for cows in the second lactation 33 to 60 and 45 to 72 were considered for cows in the third parity. Gaines, Wood, Wilmink and MilkBot Model were used to fit lactation curves. The NLIN procedure of SAS9.4 software and Gauss-Newton algorithm were used to fit lactation curves. The AIC criterion was used to select the best model. After the parameters of the curve for the animals were individually fitted using the best model. The parameters estimated for the animals were used as traits to estimate the (co)variance components and genetic parameters. A multi-trait animal model was used to estimate (co)variance components. To estimate the (co)variance components, the restricted maximum likelihood method and AI algorithm used. Airemlf90 software was used to estimate variance components.
Results and Discussion: The MilkBot model was chosen as the best model for all three lactations. It can be seen that cows with the first Lactation reached the peak of production later and also have better milking Persistency and these results are consistent with the physiology of cows and the production process of different lactation cows. Estimated heritability for the cows of the first lactation is equal to 0.107 (0.040), 0.052 (0.020), 0.034 (0.010) and 0.019 (0.020) respectively. The second is equal to 0.110 (0.050), 0.014 (0.010), 0.029 (0.020) and 0.086 (0.011) and for the cows of the third Lactation is 0.123, (0.010) 0.078, (0.051) 0.026 and (0.020) 0.0631. The genetic correlation between parameters of lactation curve was obtained in the range of -0.085 between parameters a and d of the first Lactation cows and 0.891 between parameters a and b. Phenotypic correlation was obtained in the range of 0.119 between b and c parameters of the second Lactation cows and 0.697 between a and c parameters of the third Lactation cows. In general, the heritability’s obtained for the lactation curve are low and only first parameter has a relatively moderate heritability. The small heritability of these traits indicates that the lactation curve has been greatly influenced by environmental factors and as a result of the genetic selection of these traits there has been little genetic progress and it takes many generations to reach an optimal level. One of the possible ways to improve the lactation curve is to examine the traits that have a high genetic correlation with the parameters of the lactation curve and have suitable heritability, which can be used as a suitable criterion to improve the shape of the lactation curve. The range of genetic correlations is between -0.085 to 0.891 and the range of phenotypic correlations is between 0.119 and 0.697. The genetic correlation between parameter a and parameters b, c and d show that the highest genetic correlation is between parameter a and b and the phenotypic correlation is between parameter a and parameter c. The positive correlation between initial production and the increasing slope to peak production indicates that cows with higher initial production reach peak production at a suitable fast.
Conclusion: Based on the findings of this research, the MilkBot model was chosen as a very efficient model in fitting the lactation curve and also estimating important parameters in the management and genetic improvement of herds. The heritability of the parameters of the lactation curve shows the low effect of the additive genetic effect on the shape and structure of the curve, and it can be concluded that genetic factors are less effective on the curve than other factors such as environmental factors, and as a result, a large number of generations is needed for the genetic improvement of the herd. It is necessary in terms of the lactation curve. Genetic correlation of lactation curve parameters with other traits can be an important and appropriate way to improve the genetics of lactation curve by selecting correlated traits.