Prediction of the effect of enzymes on chick performance when added to cereal-based diets: Use of a modified log-linear model

A previous study demonstrated that a log equation could be used to predict the relationship between the amount of a crude enzyme added to a diet and c hick performance. The objective of the current study was to determine if a modification of the original equation, in conjunction with a computer progr am, would overcome some of its limitations. The modified equation was Y = A + B log (CX + 1), where Y is the estimated performance value; A is the int ercept that represents the performance without enzyme supplementation; B, t he slope of the equation (performance change per log unit of an enzyme in t he diet), is a measure of an enzyme efficacy; C is an amplified factor; and X is the amount of enzyme in the diet. The results demonstrated that the n ew model more accurately predicted chick performance than that of the origi nal equation with correlations (r) between chick performance and amount of different enzymes added to the diet ranging from r = 0.80 to 0.99 (P < 0.05 ). In addition, the same trends were found when the model was used to asses s the efficacy of a given enzyme added to corn-, wheat-, barley-, and rye-b ased diets or for combinations of two dietary components (rye and wheat). T he model proposed in this study provides a new means of assessing the overa ll efficacy of an enzyme preparation. This model could be routinely used by enzyme and livestock producers to establish the best combination of differ ent cereals and enzymes so as to maximize net returns.