Genetic relationship among somatic cell score and some milking traits in Holstein-Friesian primiparous cows milked by an automated milking system

Abstract

The automated milking system provides breeders with a large amount of automatically collected information about each cow in herd that cannot be easily obtained in non-robotised systems. This knowledge can be used in breeding programs improving somatic cell count (SCC) level. The objective of this study was to estimate heritabilities and genetic correlations among test-day (TD) somatic cell score (SCS) and selected milking traits, such as daily milk yield (MY), milking frequency (MF), milking time (MT) and milking speed (MS), attachment time (AT) to single teat cups, electrical conductivity (EC) and milk temperature (MTEMP). Data were collected for 1899 Polish Holstein-Friesian primiparous cows milked in an automatic milking system. Genetic parameters of the studied traits were estimated using Bayesian method via Gibbs sampling and two-trait random regression animal model with fixed effect of herd x TD, fixed regressions on days in milk (DIM) nested within age at calving by season of calving and RR for additive genetic and permanent environmental effects. Both fixed and RR were fitted with fourth-order Legendre polynomials on DIM. The estimated daily heritabilities were in the following ranges: MY – 0.162–0.338, MF – 0.156–0.444, MT – 0.090–0.320, MS – 0.252–0.665, AT – 0.105–0.394, EC – 0.269–0.466, MTEMP – 0.135–0.304 and SCS – 0.155–0.321. The heritabilities for traits expressed on a 305-d basis were moderate to high: 0.460 for MY, 0.514 for MF, 0.315 for MT, 0.431 for MS, 0.256 for AT, 0.386 for EC, 0.407 for MTEMP and 0.359 for SCS. Genetic correlations between traits on a 305-d basis showed that SCS was most strongly genetically correlated with MTEMP (0.572) and MS (0.480), whereas genetic relationships of SCS with MT (0.221) and EC (−0.216) were moderate. Phenotypic correlations between traits on a 305-d basis were moderate or low. Somatic cell score was negatively phenotypically correlated with MY, MF and MT, with the highest relationship with MT (−0.302). The largest positive phenotypic correlations were observed between SCS and MS (0.311) and with MTEMP (0.286). In summary, it is concluded that there is a chance to carry out effective selection for lower SCS and for some other traits, in particular MS and MTEMP. The obtained results are promising enough to conduct further research to evaluate how these traits can be used both to increase the accuracy of genetic evaluations of SCC and to improve udder health.