Abstract #122
Section: Ruminant Nutrition (orals)
Session: Ruminant Nutrition 1: Protein and Amino Acid I
Format: Oral
Day/Time: Monday 11:30 AM–11:45 AM
Location: Junior Ballroom D
Session: Ruminant Nutrition 1: Protein and Amino Acid I
Format: Oral
Day/Time: Monday 11:30 AM–11:45 AM
Location: Junior Ballroom D
# 122
Modeling milk protein yield responses to amino acid supply of dairy cows fed silage-based diets.
A. Vanhatalo1, T. Kokkonen1, P. Huhtanen*2, 1University of Helsinki, Helsinki, Finland, 2Swedish University of Agricultural Sciences, UmeƄ, Sweden.
Key Words: milk protein yield, amino acid supply, modeling
Modeling milk protein yield responses to amino acid supply of dairy cows fed silage-based diets.
A. Vanhatalo1, T. Kokkonen1, P. Huhtanen*2, 1University of Helsinki, Helsinki, Finland, 2Swedish University of Agricultural Sciences, UmeƄ, Sweden.
Production responses to changes in nutrient supply in dairy cows can be predicted with a reasonably accuracy from a large data set of milk production studies using a mixed model regression analysis. In this study, we extended this approach to a level of using individual AA supply for predicting milk protein yield (MPY). Our hypothesis was that the predictions of MPY could be improved by including the supply of individual AA supply compared with the basal model based on ME and Feed-MP supply. Treatment mean data were collected from feeding trials with cows fed silage-based diets supplemented with concentrates. The total data set included 1102 diets in 246 studies. The ME and MP supplies were calculated according to the Finnish system. For calculating the individual AA supply from RUP, the AA profiles were adopted from tabulated values. Microbial AA profile was derived from studies in dairy cows fed silage-based diets. PROC MIXED of SAS was used to develop models predicting MPY. Linear and quadratic effects of the supply of individual AA (g/kg MP or g/kg EAA) were included in the basal model with ME intake and supply of Feed-MP as independent variables. Variation in estimated supply of AA was small (mean CV 2.3% for EAA/MP) .The basal model predicted MPY well with adjusted root mean squared error 16.8 g/d. Individual AA (g/kg MP) only marginally improved predictions. The effect was significant for Arg (P < 0.01), His (P = 0.04) and Leu (P < 0.001), but quantitatively the effects were small. Quadratic effects were significant for Leu and Met with maximum MPY at 79 and 25 g/kg MP, respectively. The effects of Ile, Thr and Val were negative (P < 0.01) when included as a second AA with Leu. When AA were expressed as g/kg EAA Arg (P < 0.01) and Leu had positive (P < 0.001) effects on MPY, whereas the effects of Ile, Thr and Val were negative (P < 0.001). Quadratic effects were significant for Leu and Met with maximum MPY at 174 and 53 g/kg EAA, respectively. The models imply that potential for increasing MPY of cows fed silage-based diets is small reflecting balanced AA profile from microbial protein and RUP.
Key Words: milk protein yield, amino acid supply, modeling