Testing an approach to account for daughter misidentification in the estimation of breeding values — ASN Events
  1. Schedule
  2. Concurrent 17 - Industry focus - II
  3. Testing an approach to account for daughter misidentification in the estimation of breeding values

Testing an approach to account for daughter misidentification in the estimation of breeding values (#99)

Nicolas Lopez-Villalobos 1 , Jeremy R Bryant 2 , Bevin L Harris 3 , Dorian J Garrick 4
  1. Massey University, PALMERSTON NORTH, MANAWATU, New Zealand
  2. DairyNZ Limited, Hamilton, Waikato, New Zealand
  3. Livestock Improvement Corporation, Hamilton, Waikato, New Zealand
  4. Department of Animal Science, Iowa State University, Ames, IA, USA

A stochastic simulation was developed to account for daughter misidentification (DaMi) in the estimation of breeding values (EBV) of bulls. Miss-assignment of parentage was simulated in 20% of daughters of 50% of bulls in 30 of 40 commercial herds. A bias of -40 kg protein was assigned to those cows that were misidentified as heifer calves. The model for genetic evaluation included fixed effects of contemporary group and random effects of additive animal, permanent environment and residual. The approach to account for DaMi was to include a regression coefficient for each sire that reflected DaMi. Compared to PT EBVs, parent average (PA) EBVs were overestimated by 2.3 and 3.4 kg protein for bulls with and without DaMi. Compared to PT EBVs, RP EBVs were underestimated by 3.5 kg and overestimated by 5.1 kg protein for bulls with and without DaMi. The model that accounted for sires with DaMi removed biases from PA to PT for both groups of bulls, but compared to PT EBVs, RP EBVs were underestimated by 1.0 kg protein for both groups of bulls. This set of simulations indicates that paternity-verified status for a sire-daughter record can remove biases in genetic evaluation caused by DaMi.

Full Paper