Today's seedstock herds have usually gathered data for a
number of years and for multiple traits. In addition, they may wish to combine
data across multiple breeding sites or breeds in a single analysis. Therefore,
genetic evaluation systems like PIGBLUP must be able to handle large data sets
efficiently. A data set containing 295,000 animals with 795,000 production and
126,000 reproduction records was analysed using PIGBLUP. The data included 5
breeds and 7 user-defined management groups. The two multivariate analyses included
9 production traits and 3 reproduction traits. The analysis of this data set
took 25 minutes to complete on a laptop computer equipped with a Pentium III
processor running at 1.2 GHz and 256 MB RAM. The analytical black box of PIGBLUP
has been specifically developed for pig breeding programs and the program is
able to analyse large data sets with hundreds of thousands of records within
minutes.
