Abstract
In dairy cattle breeding,health and fertility traits have recently been included in a large number of national breeding goals.The effectiveness of breeding decisions and management changes to improve health and fertility possibly interact through genotype × environment interaction (G×E). G×E is a phenomenon in which different genotypes respond differently to changes in an environment. It can consist of the following effects: heterogeneous genetic variances across environments, genetic correlation of a trait expressed in different environments being less than 1.0 (reranking), and heterogeneous genetic correlations between traits across environments. In this thesis, G×E for health andfertiltity, as well as for yield, has been investigated using reaction norm models. In the reaction norm models, breeding values and genetic parameters were modeled as a function of an environmental descriptor using random regression. The dimensions of the model were expanded from linear random regressions to higher order random regressions, to include two parameters to define the environment, and to multivariate reaction norm models.Many environmental descriptors were investigated in this thesis, such as production level, farm size, average somatic cell score and calving interval, however, it appears that the herd parameters linked to nutrition and energy balance are most important for G×E.Significant G×E was detected in 86% of the situations for yield traits, but only in 14% of the situations for health and fertility traits, although significantrerankingwas found for SCS, mastitis and survival. Estimated G×E effects mainly consisted of heterogeneous genetic variances with limitedreranking. As a result of heterogeneous variances in different traits, the relative importance of fertility compared to yield doubled across environments. Estimated G×E effects for SCS indicated morererankingof animals based on analysis of test day records, than on lactation averages. It was shown that selection for increased yield is expected to lead to increased environmental sensitivity for yield, while selection for better fertility is expected to lead to decreased environmental sensitivity for fertility. The models presented in this thesis can be used to account for the effect of herd environment on a trait and the relations between traits, and therefore enable to make accurate predictions of breeding values across environments.
Original language | English |
---|---|
Qualification | Doctor of Philosophy |
Awarding Institution |
|
Supervisors/Advisors |
|
Award date | 17 Mar 2006 |
Place of Publication | [S.l. ] |
Print ISBNs | 9789085043751 |
DOIs | |
Publication status | Published - 17 Mar 2006 |
Keywords
- dairy cattle
- genotype environment interaction
- milk yield
- animal health
- fertility
- breeding value
- genetic parameters
- mathematical models
- animal breeding