## Parametry charakteryzujące kształt krzywej laktacji jako cechy pomocne przy szacowaniu wartości hodowlanej bydła

### Maciej Gierdziewicz

#### Abstract

Increasing milk yield of dairy cows enforces searching compromise between daily yields, especially peak yield, and lactation length. At present, linear test-day models have been developed for estimating dairy cattle breeding values. Consequently, all test-day records are used directly in genetic evaluation models. This allows the mean and shape of the lactation curve to differ between cows. Given such a flexible methodology, individual lactation curves can be calculated. The widely exploited advan¬tage of this approach is the ability to estimate lactation milk yield more precisely. The interesting and less frequently mentioned benefit is that knowledge of lactation curve parameters connected, for example, with its maximum or inflection point, could help detect incorrect curve shape and enable influencing that shape in the next generations via selection. Ten traits of lactation curve shape for milk traits have been defined: day of peak yield (D2), day of reaching the inflection point after the peak (D3), initial yield (W1), peak yield (W2), yield at the inflection point after the peak (W3), initial rate of growth of the daily yield (TWW1), growth rate (negative value) at the inflection point (TSW3), time interval between the peak and the following inflection point (C23), difference (positive value) between peak yield and the yield at the inflection point (SW23), rate of decline in yield (positive value) between the lactation peak and the inflection point (SSW23). The aim of the work was to estimate heritabilities, animal breeding values and genetic trends for 10 traits characterizing the lactation curve, related to its starting point, the peak and the inflection point, basing on test day yields of Polish Holstein-Friesian cows kept in cowhouses of HZZ Osowa Sidi, for the purpose of possible use of these 10 characteristics of the lactation curve as additional traits in breeding value estimation of dairy cattle. The data from nine cowhouses of HZZ Osowa Sidi were milk, fat and protein yields from 157761 test-day records of 14523 lactations of 6359 cows born from 1989 to 2010. The cows were daughters of 663 sires and 4703 dams. Average daily milk, fat and protein yield was 27.32 kg, 1.05 kg and 0.93 kg, respectively. Three linear models were fitted to the data. The first linear model fitted to the testday records contained fixed regressions on days in milk. In three variants of this model three lactation curve models were used: Ali-Schaeffer model, Legendre 4th degree polynomials and Wood model. The graphs with lactation curves for milk production traits with most pronounced differences between classes of fixed effects were presented. The second linear model included the same fixed regressions as the first one and, additionally, random regressions on individual lactation curve parameters, both for permanent environment effect and for additive genetic effect. For each of three milk production traits and for each of three lactation curve mod¬els, 10 traits characterizing the shape of the individual lactation curve were esti¬mated for each cow. The third model fitted to each of the 10 traits mentioned before, was an individual animal model with fixed effects, permanent environment effect and additive genetic effect (animal effect). The model was used to estimate variance components and, consequently, heritabilities of 10 lactation curve traits, breeding values of animals, and genetic trends of 10 traits mentioned before. Fitting the first linear model let detect differences between lactation curves for the first and later lactations. Also, the curves estimated with Legendre polynomials differed from those calculated with the use of other models. The quality of fitting the lactation curves obtained from the second model was assessed with R2 coefficient the values of which ranged from 0.62 to 0.77. Among the curves for milk yield calculated from Ali-Schaeffer model and from Legendre polynomials, 23% atypically shaped curves were detected. This is the reason for not obtaining the values of all 10 lactation curve traits for all cows. For 77% lactation curves all traits were calculated. Average values of 10 traits ranged as follows, depending on the model of the lactation curve: day of peak yield (D2) from 53 to 64, day of reaching the inflection point after the peak (D3) from 111 to 187, initial yield (W1) from 24 to 30 kg, peak yield (W2) from 34 to 37 kg, yield at the inflection point after the peak (W3) from 28 to 34 kg, initial rate of growth of the daily yield (TWW1) from 0.13 to 1.80 kg • d-', growth rate at the inflection point (TSW3) from —0,08 to —0,07 kg . d-1, time interval between the peak and the following inflection point (C23) from 76 to 127 days, difference (positive value) between peak yield and the yield at the inflection point (SW23) from 4.7 to 7.0 kg, rate of decline in yield between the lactation peak and the inflection point (SSW23) from 0.05 to 0.06 kg . d-1. Lactation curves for fat yield resembled straight lines, day of peak yield ranged from 53 do 109, day of reaching the inflection point (D3) from 65 to 351, average yields ranged: WI from 0.88 to 1.10 kg, W2 from 0.98 to 1.15 kg, W3 from 0.69 to 1.15 kg, rate of decline in yield between the lactation peak and the inflection point (SSW23) from 0.001 to 0.003 kg • d-1. Lactation curves for protein yield resembled those for milk yield, however, a little more atypical shapes have been detected among the curves for protein yield. The estimated heritabilities for the lactation curve shape traits of, presumably, the greatest importance (TWW1, D2, W2, TSW3 and SSW23) for milk yield were, correspondingly: 0.27-0.65, 0.22-0.29, 0.15-0.25, 0.34-0.40 and 0.34-0.41. For protein yield they were, respectively: 0.10-0.54, 0.11-0.18, 0.33-0.36, 0.22-0.30 and 0.31-0.41. Linear genetic trends with only satisfactory level of the determination coefficient were detected only for a few lactation curve traits that could have been subject to indirect selection before. However, all 10 lactation curve traits considered in this work are characterized by genetic variability large enough to perform selection for those traits in future. The traits of lactation curve shape such as initial rate of growth of the daily yield (TWW1), day of peak yield and peak yield (D2 and W2), growth rate at the inflection point (TSW3) and rate of decline in yield between the lactation peak and the inflection point (SSW23) could be useful in selection for an appropriate shape of the lactation curve, depending on the breeding strategy (annual production cycles versus extended lactations). From the results of this work and from the literature it can be concluded that lactation curve traits, in particular those describing the peak and the inflection point, could constitute valuable auxiliary criteria in selection of dairy cattle.Other language title versions | Parameters characterizing lactation curve shape as auxiliary traits in estimating cattle breeding value |

Book type | Monograph |

Other categories | ROZPRAWA_HABILITACYJNA |

Author | |

Publisher | Uniwersytet Rolniczy im. Hugona Kołłątaja w Krakowie, MNiSW [80] |

Publishing place (Publisher address) | Kraków |

Issue year | 2015 |

Book series /Journal (in case of Journal special issue) | Zeszyty Naukowe Uniwersytetu Rolniczego im. Hugona Kołłątaja w Krakowie. Rozprawy, ISSN 1899-3486, (0 pkt) |

Vol | 529 |

No | 406 |

Pages | 124 |

Publication size in sheets | 8 |

Language | pl polski |

Score (nominal) | 25 |

Citation count* |

* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.

Back