Search

Six multiparous Holstein cows were used in a 6 x 6 Latin square to investigate the ability of the Cornell Net Carbohydrate and Protein System to predict accurately rumen microbial yield, plasma urea N, and milk urea N. Estimations for microbial protein yield were compared with the measured excretion of purine derivative N in urine. A 3 x 2 factorial arrangement of treatments was adopted. Three concentrations of a rumen-undegradable protein (RUP) supplement (4.5, 14.9, and 29.1% of dry matter intake) and two levels of feed restriction (90 and 80% of ad libitum intake) were the corresponding factors. No effect of concentration of RUP supplement or feed restriction was detected on the excretion of purine derivative N in urine (mean, 18.5 g/d). Conversely, the Cornell system predicted a linear decrease in metabolizable protein from bacteria as the concentration of the RUP supplement increased. The Cornell system also predicted a significant reduction in metabolizable protein of microbial origin as feed restriction was increased. Measured values and values derived from the Cornell system for plasma and milk urea N increased linearly as the concentration of the RUP supplement increased. The Cornell system overpredicted milk urea N for cows consuming the highest RUP concentration. Predictions by the Cornell Net Carbohydrate and Protein System were of limited value because the empirical nature of the model is insufficiently rigorous to yield accurate predictions under the conditions described herein.

An experiment was designed to determine the response of milk protein production and N utilization in dairy cows to supplementation of a predominantly rumen-undegradable protein (RUP) mixture with a fixed amino acid (AA) pattern and the response to the amount of feed intake. The experiment was designed as a 6 x 6 Latin square with a 3 x 2 factorial arrangement of treatments. The factors were three concentrations of RUP supplement (4.5, 14.9, and 29.1% of dry matter intake) and two levels of feed intake restriction (10 and 20%) of the basal diet. The supplement was designed to approximate a postruminal AA pattern that was similar to bovine caseins for Met, Lys, Phe, His, and Thr. Measurements were made during the last 5 d of each 21-d period. Milk protein production responded linearly as the concentration of RUP supplement in the treatment diet increased within the given range. The difference in feed intake restriction did not affect milk protein production. Efficiency of N utilization for milk production exceeded 30% for cows fed the lowest RUP supplement. Results indicated that there is an opportunity to increase milk protein production by using RUP formulations that are balanced for AA while minimizing waste N excretion.

A dataset of 177 individual nitrogen balances from dry and lactating cows was split in two independent groups: training dataset (n = 130) and challenge dataset (n = 47). The training dataset was used to develop multiple linear regressions (MLR) and artificial neural networks (ANN) aimed at predicting the urinary excretion of total (NURI) and that of purine derivative nitrogen (PDN). Input variables for the prediction of NURI were crude protein (CP) intake, effective degradability of non-protein dry matter (DM), neutral detergent fiber (NDF) content of the diet, live weight and milk yield. Live weight, total carbohydrate intake, the ratio of non-protein DM degraded to CP degraded and milk yield corrected for DM intake were entered to predict PDN. The regression between predicted and observed values for the training dataset showed a better statistical accuracy of ANN than did MLR models, especially for PDN. The evaluation of the two models on the challenge dataset showed similar determination coefficients, either when predicting total nitrogen excretion (0.623 and 0.614 for ANN and MLR, respectively) or PDN (0.688 and 0.666, for ANN and MLR, respectively). Moreover, both approaches were affected by a tendency to under-predict both targets at high levels of NURI and PDN. However, with the ANN approach, it is possible to study the response of the model to modifications of individual inputs by the so-called response analysis. This unique feature could be used to study the effect of different physiological situations as well as providing hypotheses for additional research. [ABSTRACT FROM AUTHOR]