New Test Day Model for the Genetic Evaluation of mastitis in dairy cattle

Authors

• E. Negussie
• K. Vuori
• I. Strandén
• E. A. Mäntysaari MTT Agrifood Research, Biotechnology and Food Research, Biometrical Genetics, 31600 Jokioinen, Finland

DOI: https://doi.org/10.33354/smst.77002

Keywords:

Dairy cattle, genetic evaluation, mastitis resistance, random regression model

Abstract

In this study, genetic parameters of test-day (TD) somatic cell score (SCS) and lactation average (LA)
clinical mastitis (CM) were estimated using a random regression model (RRM) that combine two different
data models. A multitrait RRM (mt-RRM) was then developed for the genetic evaluation of mastitis.
Estimates of breeding values (EBVs) from the mt-RRM were compared to corresponding multitrait LA
model (biv-LAM) and univariate LA models (univ-LAM). A total of 147500 and about 5.6 million records
from 27500 and 1.4 million Finnish Ayrshire cows were used for estimation of genetic parameters and
prediction of breeding values, respectively. Heritabilities of CM1 and CM2 traits: (CM1, -7 to 30 and
CM2, 31 to 300 DIM) were 0.026 and 0.016, respectively, while for TD SCS they ranged from 0.06 to
0.11. During first lactation, the genetic correlations between TD SCS and CM1 and between TD SCS and
CM2 varied from 0.40 to 0.77 and from 0.34 to 0.71, respectively. In genetic evaluation of mastitis, model
comparisons have showed that mt-RRM has high model predictive ability and high standard deviation of
breeding values. Moreover, it has added advantages of making efficient use of available TD SCS
information and offers proofs for bulls and cows. Therefore, mt-RRM can be used as best practical model
in the future evaluation of animals for mastitis resistance.