Estimates of genetic parameters for traits measured in Phase A, C and D of the South African performance testing scheme

Abstract

Numerous growth, reproduction, feed efficiency and shoulder measurement traits are recorded on beef cattle under the auspices of the South African Beef Cattle Performance Testing Scheme (NBCPTS). Best Linear Unbiased Prediction (BLUP) was first introduced to the South African beef industry on a national level in 1994, when univariate animal models were used to predict breeding values for traits measured in Phase A (birth and weaning weight) and B (yearling and 18 month weight) of the Scheme (Van der Westhuizen et al., 1995). In 1996 traits recorded in Phase D (growth tests on-farm) were also included into the analyses. Traits evaluated in Phase D include Average Daily Gain (ADGD), Kleiber Ratio (KR), Scrotum Circumference (SCD) and Shoulder Height (SHD). Although an univariate analysis uses all available pedigree information to calculate EBVs, it does not account for selection between traits, nor does it use the association (correlations) between traits to increase the accuracy of the EBVs. A common management practice in South Africa is to cull poor performing and functionally inefficient animals at weaning before further testing in Phase D, or alternatively, animals with above averageperformance are sent to central performance testing stations (Phase C). There is thus a high level of selection at this stage which should be accounted for in the national genetic evaluation. Traits measured in Phase C tests include Average Daily Gain (ADGC), Food Conversion Ratio (FCRC), Scrotum Circumference (SCC) and Shoulder Height (SHC). The objective of this study was to estimate covariance components between weaning weight and traits measured in Phase C and D of the NBCPTS. Results from this study will allow Phase C data to be included into the routine genetic evaluations, will account for the selection that occurs at weaning when analysing Phase C and D data, and will also increase the accuracy of EBVs by using the correlations between traits.