Submitted to: Journal of Dairy Science
Publication Type: Abstract Only
Publication Acceptance Date: 3/23/2004
Publication Date: 7/25/2004
Citation: Norman, H.D., Wright, J.R., Powell, R.L., VanRaden, P.M. 2004. Merit of obtaining genetic evaluations of milk yield for each parity on Holstein bulls [abstract]. Journal of Dairy Science. 87(Suppl. 1):2. Interpretive Summary:
Technical Abstract: Modeling separate Predicted Transmitting Abilities (PTA) for each parity is becoming more frequent internationally. A concern when using a single trait genetic evaluation model is that changes in PTAs could be larger than expected for bulls whose daughters deviate considerably from the typical response due to aging. The objective of this research was to examine how difference in rate of maturity of bulls' daughters impacts changes in bull evaluations across time. The 305-d milk records for Holsteins first calving between 1960 and 1998 were used to calculate 12 tailored PTAs for each bull. Tailored PTAs (controlled by parity combinations and run dates) were calculated for 1996 to1999 run years using records of cows first calving prior to each January. Four annual evaluations included only first lactations on daughters and contemporaries (PTA1). Four more included the first two lactations available (PTA1-2), and the last four included the first three lactations (PTA1-3). The three PTAs (different parities) for each year included cows first calving by the same cutoff date. Tailored PTAs had more uniformity across years in mean records/daughter than published PTAs. Stability (absolute differences, standard deviations (SD) of differences, and correlations between bull evaluations) across time was compared. Bulls born after 1984 having >=500 daughters were examined because their high precision and recent activity. SD of bull differences in PTA1, PTA1-2, and PTA1-3 for milk between 1996 and 1997 were 28, 28, and 27 kg compared to 63 kg for the published evaluations. Similarly, SD between 1996 and 1999 were 36, 32, 32, and 79 kg, respectively. PTA1-2 was usually intermediate to PTA1 and PTA1-3 for any bull having a large difference between PTA1 and PTA1-3. Results suggest that a modification to the current evaluation model should reduce the fluctuations in PTAs for individual bulls across time and should improve the accuracy of evaluations.