Cluster analyses to explore the genetic curve pattern for milk yield of Holstein

Abstract

Animal selection in dairy cattle can vary depending on the objectives of the breeding programs. The objective of this study was to explore the genetic curve pattern of EBVs for test day milk yields (TDMY) in Holstein cows using cluster analyses to identify the most suitable animals for selection based on their genetic curve for milk yield. A data set with 29,477 monthly TDMY records from 3543 first lactations of Brazilian Holstein cows were used to predict the breeding values for TDMY with random regression model. Hierarchical and non-hierarchical cluster analyses were performed based on the EBVs for 30, 60, 90, 120, 150, 180, 210, 240, 270, and 305 days in milk (DIM) to explore the genetic curve patterns of milk production of animals within the population. At first moment, the population was divided into three groups based on animals' genetic curve pattern for milk yield using hierarchical cluster analysis. According to non-hierarchical cluster analysis, one of those groups had EBVs along the lactation curve above the population average. Further cluster analysis done only with those animals with genetic curve pattern above the population mean showed specific subgroups of animals with different genetic curves for milk yield despite of all of those animals had EBVs above the population average, along the lactation curve. It indicated that specific subgroup of animals with a specific genetic curve pattern for milk yield can be chosen depending on the objectives of the breeding program. It was concluded that the cluster analyzes could be used to select animals based on the shapes of the genetic curve for milk production together with the EBV for milk yield at 305 days in milk. Thus, it can be possible to select at the same time more productive animals with genetic curves that met the goals of breeding programs that take into account the milk production in other parts along the milk production curve.