Fit of Wood's function to daily milk records and estimation of environmental and additive and non-additive genetic effects on lactation curve and lactation parameters of crossbred dual purpose cattle

The objectives of this study were to estimate the environmental and additive and non-additive genetic effects on lactation curve and lactation parameters of crosses of Holstein (H), Brahman (B) and Brown Swiss (BS) in Olancho region in Honduras. The data consisted of 54,517 milk yield records from 192 dual-purpose crossbred cows lactating from 2000 to 2005 at the Universidad Nacional de Agricultura de Honduras (UNA). The lactation curve and lactation parameters of interest were the scaling factor to represent yield at the beginning of lactation (a). the factor associated with the inclining (b) and declining (c) slopes of the lactation curves, and the milk yield at initial day of lactation (MY20), peak milk yield (MYmax), day at peak milk yield (t(max)), and the total milk yield (TM) per lactation, respectively. The incomplete gamma function (Wood function) was used to estimate lactation curve and lactation parameters from daily milk records of H x B, H x BS and BS x B crossbred cows. The environmental, additive and non-additive genetic effects on lactation curve and lactation parameters were estimated using Dickerson and Kinghorn models. The coefficients of determination of fitness of Wood's function (R-2) ranged from 80% to 97% with an average of 93%. The lactation curve of the crossbred cows was similar to those reported for dairy cows grazing in the tropics. Lactation parameters such as MYmax t(max) and TM were significantly (P<0.05) influenced by environmental sources of variation suggesting the necessity of differential management strategies. The moderate to large positive phenotypic correlation of MYmax and TM indicate that one of the milk yield parameters could be used as a selection criterion to improve either one or both traits. Despite the fact that both genetic models showed similar patterns, the absolute value of the parameters varied. For both models, individual additive genetic breed effect for H breed were significant (P<0.05) and contributed more to TM than the BS breed. In the Dickerson model, highly positive significant (P<0.01) effect on TM for HxBS and BSxB crosses was found. The Kinghorn model did not show significant effects of dominance on this parameter. The estimate of recombination effect for all crosses involving B breed were negative and significant (P<0.05) for positive correlated lactation curve parameters. Although the inclusion of nonadditive effects on crossbreeding genetic effects were not all significant for lactation curve and lactation parameters, non-additive effects should be taken into account to improve the Honduran dairy cattle production management. (C) 2009 Elsevier B.V. All rights reserved.