Analysis of the linear models using to assess the influence of different factors on the productivity of dairy cows

UDC 636.2.034/57.087

О. Kramarenko,
O. Potryvaieva

To determine the influence of various factors on the milk yield of cows, we used a linear model of a mixed type, as a random factor, the genotype of a bull-producer, and as fixed – the number of lactation, the year of birth and the calving season of cows. In total, data on 526 lactations per 113 red steppe cows were included in the analysis, which were kept in the conditions of State Enterprise “Plepreproductor” Stepnoe “(Ukraine, Mykolayiv region) during 2001-2014.
Data on milk yield for ten months (M1-M10) and 305 days of lactation (Y305) were used as a dependent variable. For individual gradations of factors, the estimates of the coefficients with the employer of the LSM procedure (Least squares means ± SE) were computed in the model, and the level of reliability of their deviation from zero (p-estimation) was determined. All calculations were carried out using the module “GLM” (General Linear Model) of the MiniTab software package v. 15.
We have established an insignificant, but it is possible that some bulls-producers influence the hopes of their daughters during M2 … M4. Thus, it was found that the milk of the Altea was lower for the M2-M3 daughters, whereas the milk yield for M2 … M4 in the daughters of the bull of Narcissus was, on the contrary, higher than the average population value. In addition, the last almost 130 kg prevailed in relation to milk yield for 305 days of lactation.
The first-borns were characterized by less milking during the first half of the lactation (especially, according to M1-M2), and by 416 kg inferior to the average population estimate. On the other hand, cows at the age of eighth lactation, on the contrary, showed significantly higher milk yield during the average lactation months (M4 … M8), and an average population estimate prevailed on 662 kg of milk.
In addition, a high level of milk productivity was established for cows born in 2005, 2007 and 2009, which is primarily due to the increase in their milk productivity in the first half of lactation.
The presence of a highly probable effect of the calving season on the level of milk productivity of cows is manifested primarily in the increase in milk yields (especially in M4 … M8) in January calving, and, conversely, in decline – with calving in June and July.

Keyword: Linear models, number of lactation, year of birth, season of calving, dairy cattle.

Analysis of the linear models using to assess the influence of different factors on the productivity of dairy cows.

1. Getya A. A., Dodenkhoff Y. Zastosuvannia BLUP-metodu pry orhanizatsii otsinky selektsiinoi tsinnosti svynei v Ukraini / A. A. Getya, Y. Dodenkhoff // Tekhnolohiia vyrobnytstva i pererobky produktsii tvarynnytstva. – 2010. – Vypusk 3 (72). – S. 52-55.
2. Danshin V. A. Ocenka geneticheskoj cennosti zhivotnyh / V. A. Danshin – K. : Agrarna nauka, 2008. – 180 s.
3. Sovremennye metody geneticheskogo kontrolya selekcionnyh processov i sertifikaciya plemennogo materiala v zhivotnovodstve / [N. A. Zinov’eva, P. M. Klenovickij, E. A. Gladyr’ i dr.]. – M. : RUDN, 2008. – 329 s.
4. Kuznecov V. M. Metody plemennoj ocenki zhivotnyh s vvedeniem v teoriyu BLUP / V. M. Kuznecov. – Kirov : Zonal’nyj NIISKH Severo-Vostoka, 2003. – 358 s.
5. Kramarenko S. S. Vykorystannia liniinykh modelei (BLUP) dlia otsinky pleminnoi tsinnosti koriv za molochnoiu produktyvnistiu / S. S. Kramarenko, O. I. Potryvaieva // Visnyk ahrarnoi nauky Prychornomoria. – 2016. – Vyp. 2(2). – S. 187-192.
6. Kramarenko S. S. Novi metody matematychnogho modeljuvannja laktacijnykh kryvykh za dopomoghoju interpoljaciji / S. S. Kramarenko // V kn. : Materialy mizhnarodnoji naukovo-praktychnoji konferenciji «Novitni tekhnologhiji skotarstva u ХХІ stolitti» (Mykolajiv, 4-6 veresnja 2008 r.). – Mykolajiv : MDAU, 2008. – S. 159-164.
7. Ryan B. F. MINITAB Handbook : Update for release 16 / B. F. Ryan, B. L. Joiner, J. D. Cryer. – Pacific Grove, CA, USA: Brooks/Cole Publishing Co., 2012. – 560 p.
8. Wood P. D. P. The relationship between the month of calving and milk production / P. D. P. Wood // Animal Science. – 1970. – V. 12. – №. 2. – P. 253-259.
9. Miller P. D. Joint influence of month and age of calving on milk yield of Holstein cows in the northeastern United States / P. D. Miller, W. E. Lentz, C. R. Henderson // Journal of Dairy Science. – 1970. – V. 53. – №. 3. – P. 351-357.
10. Ray D. E. Season and lactation number effects on milk production and reproduction of dairy cattle in Arizona / D. E. Ray, T. J. Halbach, D. V. Armstrong // Journal of Dairy Science. – 1992. – V. 75. – №. 11. – P. 2976-2983.
11. Ptak E. Interaction of age and month of calving with year of calving for production traits of Ontario Holsteins / E. Ptak, H. S. Horst, L. R. Schaeffer // Journal of Dairy Science. – 1993. – V. 76. – №. 12. – P. 3792-3798.
12. Abate A. L. Seasonal variation of milk persistency of Kenana× Friesian crossbred dairy cows under confinement feeding in a hot environment / A. L. Abate, M. Atta, R. N. Anthony // Animal Science Journal. – 2010. – V. 1. – №. 1. – P. 13-18.

Issue 3 (95), 2017

Estimation of the dairy cow’s breeding values using the linear models (BLUP)

UDC 636.082/57.087

S. Kramarenko,
O. Potryvaieva

In recent years, in order to estimate farm animals breeding value, more and more attention is paid to the method of BLUP (Best Linear Unbiased Prediction). The essence of this method is using of statistical adjustment for the impact of identifiable factors.
The main purpose of this study is a comparative analysis of various algorithms of linear models to obtain reliable estimates of breeding values (EBV) of animals.
As the material for the study were used zootechnical data for 113 cows of Russian Red Steppe Breed, which were keeping in State Enterprise “Plemreproductor “Stepove” (Mykolayiv region) during the 2001-2014 years. As a dependent variable were used data of the milk yield for 305 days of the first lactation.
As randomized factors was used bull-producer’s genotype. As fixed factors – year of birth, month of calving and season of calving. Finally, as the covariance were used weight in different stages of postnatal growth, exterior measurements and the age of the first effective insemination.
It was established, that the model, which includes the effect of bulls’ genotype, as well as year of cows’ birth, unfortunately, gives unsatisfactory results. Although, when considered separately, both these factors seems to has highly likely impact. So, the assessment proved to be very sensitive to the imbalance in the using of different bulls at different times.
Using exterior indicators makes only an insignificant effect in the adjustment of estimates of breeding value. The most significant effect on the estimates of breeding value of bulls had the weight of cows at the age of first lactation and the age of the first effective insemination – in this case the EBV accuracy increases by almost in a half (from 18.3% to 32.6%).

Key words: estimated breeding value (EBV), Linear models (BLUP), dairy cow.

Estimation of the dairy cow’s breeding values using the linear models (BLUP).

Issue 2(90), Part 2, 2016