نوع مقاله : مقاله پژوهشی
نویسندگان
1 بخش علوم دام، دانشکده کشاورزی، دانشگاه بیرجند
2 مؤسّسه تحقیقات علوم دامی کشور
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
Introduction: Over the past decades, a great number of mathematical models have been proposed to model the curve of the lactating dairy animals. There are two kinds of parametric and non-parametric models, among which the three-parameter Wood's gamma incomplete function has been widely used in dairy cows. Based upon the parameters of Wood's model, some lactation characteristics such as peak time, peak yield, and persistency can be calculated. The lactation features could be recognized as economic traits for which genetic selection to be practiced. Prior studies have been heavily focused on using mathematical models to calculate these lactation characteristics based on the mean of test day milk records, while they could be computed for individual quantiles of TDMR. To the best of our knowledge, no research has been undertaken so far regarding the use of quantile regression in modeling the lactation curve of Iranian dairy cows. Based on this, the main objective of the present research was to evaluate some lactation curve characteristics of Iranian dairy cows by applying the quantile regression technique.
Materials and methods: The data used in the present study were provided by the Animal Breeding Centre of Iran. The initial data set were merged and edited by SPSS software. The final data set was comprised of test day milk records (TDMR) collected from 984,690 dairy cows distributed in 579 herds from 26 provinces across the country. Total number of TDMR was 15,647,867 obtained from the first five parities of the cows calved during 1995-2021. Initially, TDMR were analysed based upon a fixed linear model fitted for different combinations of parity – calving season (20 combinations). Then, for each combination, least squares means of TDMR over the interval of 4 to 305 days in milk were utilised in Wood’s incomplete gamma function (y=atbe-ct) in a quantile manner. The Wood’s model is a parametric three-parameter (a, b, c) function that has been widely used to mathematically describe the lactation curve. The linearized form of the Wood’s model was fitted to the least square means of TDMR using the Quantreg procedure of SAS software to estimate the parameters over the quantiles of 5 to 95 by step of 5. Based on the estimated Wood’s parameters, lactation characteristics including peak time, peak yield, and persistency were calculated over these quantiles for all combinations of parity – calving season and statistical comparisons among the combination means were carried out by SPSS software.
Results and Discussion: The results of the present research indicated that peak time, peak yield as well as persistency of the lactation curve in Iranian dairy cows were not the same over different quantiles of TDMR, parities, and seasons of calving. The findings also revealed that peak yield increased as the quantile of TDMR increased. In contrast, peak time and persistency in the lower quantiles of TDMR were greater than those obtained in the upper quantiles. Over the quantiles of TDMR, minimum (50.92 d) and maximum (93.36 d) of peak time, as well as persistency (6.64 and 7.37) were observed in the cows calved in autumn and started first or third lactation period (P<0.05). Regarding peak yield, cows calved in spring and started the first lactation period had a lower mean (31.84 Kg) while the maximum mean (40.43 Kg) was observed for the cows calved in autumn and started the fourth lactation period (P<0.05). As expected, first parity cows had greater lactation persistency as compared to the other cows. Peak yield was significantly (P<0.0001) correlated with peak time (-0.9520) and persistency (-0.9516), indicating that cows with a lower level of milk yield at peak time are expected to reach peak time later and to have higher persistency after peak time.
Conclusions: The findings of the present research indicated that peak time, peak yield, as well as persistency of the lactation curve in Iranian dairy cows varied over the quantiles of test day milk records such that peak time and persistency were greater in the lower quantiles as compared to the upper quantiles. Furthermore, all above mentioned lactation characteristics are not the same over different parities and calving seasons. Based upon the results of this research, it could be suggested that these variations of the lactation characteristics should be taken into account when all records from different parities are to be used in genetic evaluation of Iranian dairy cows, which will be of great importance when test day models are to be utilized for genetic analysis of the lactation curve.
کلیدواژهها [English]
)