ARS | Publication request: GENETIC CORRELATIONS BETWEEN FIRST AND LATER PARITY CALVING EASE IN A SIRE-MATERNAL GRANDSIRE MODEL

Submitted to: World Congress of Genetics Applied in Livestock Production
Publication Type: Proceedings
Publication Acceptance Date: May 6, 2006
Publication Date: August 13, 2006
Citation: Wiggans, G.R., Van Tassell, C.P., Cole, J.B., Thornton, L.L. 2006. Genetic correlations between first and later parity calving ease in a sire-maternal grandsire model. 8th World Congress of Genetics Applied in Livestock Production. Communication 01-92.

Interpretive Summary: (Co)variance components were estimated for a multi-trait sire and maternal grandsire (MGS) threshold model where first and later parity calving ease effects were analyzed as correlated traits using Holstein data from the USA. Scores were on a 1 to 5 scale from easy or unobserved to extreme difficulty. To create datasets of manageable size, five samples of approximately 250,000 calvings were selected. These samples included the most frequently 2,600 bulls occurring as a sire or MGS. Herd years were required to have at least 20 calvings. The model included fixed year-season of calving, sex of calf within parity, and birth year groups for sire and MGS effects and random herd year, sire, MGS, and residual effects. Third and greater parities were grouped together. Estimation was by AIREML and Gibbs sampling. Correlations from Gibbs averaged 0.93 for sire and 0.80 for MGS between first and later parity. AIREML gave similar correlations. Variances for the later parity group were less than for first parity, particularly for MGS. Evaluations, particularly for MGS, can be improved by treating first and later parities as correlated traits.

Technical Abstract: (Co)variance components were estimated for a multi-trait sire and maternal grandsire (MGS) threshold model where first and later parity calving ease effects were analyzed as correlated traits using Holstein data from the USA. Scores were on a 1 to 5 scale from easy or unobserved to extreme difficulty. To create datasets of manageable size, five samples of approximately 250,000 calvings were selected. These samples included the most frequently 2,600 bulls occurring as a sire or MGS. Herd years were required to have at least 20 calvings. The model included fixed year-season of calving, sex of calf within parity, and birth year groups for sire and MGS effects and random herd year, sire, MGS, and residual effects. Third and greater parities were grouped together. Estimation was by AIREML and Gibbs sampling. Correlations from Gibbs averaged 0.93 for sire and 0.80 for MGS between first and later parity. AIREML gave similar correlations. Variances for the later parity group were less than for first parity, particularly for MGS. Evaluations, particularly for MGS, can be improved by treating first and later parities as correlated traits.