Methodologies for modelling energy and amino acid responses in poultry

The objective of this paper is to present some of the issues faced by those whose interest is to predict responses in poultry, concentrating mainly on those related to the prediction of voluntary food intake, as this should be the basis of models designed to optimise both performance and feeding programmes. The value of models designed to predict growth or reproductive performance has been improved inestimably by making food intake an output from, as opposed to an input to, such models. Predicting voluntary food intake requires the potential of the bird to be known, be this the growth of body protein or lipid, the growth of feather protein, or the rate at which yolk and albumen may be deposited daily in the form of an egg, and some of the issues relating to the description of potentials are discussed. This potential defines the nutrients that would be required by the bird on the day, which can be converted to a desired food intake by dividing each requirement by the content of that nutrient in the feed. There will be occasions when the bird will be unable to consume what is required, and predicting the magnitude of these constraints on intake and performance provides the greatest challenge for modellers. This paper concentrates on some issues raised in defining the nutrient requirements of an individual, on constraints such as high temperatures and the social and infectious environment on voluntary food intake, on some recent differences in the response to dietary protein that have been observed between the major broiler strains, and on the methodologies used to deal with populations of birds, and finally with broiler breeder hens, whose food intake is constrained by management, not by the environment. These issues suggest that there are still challenges that lie ahead for those wishing to predict responses to nutrients in poultry. It is imperative, however, that the methods used to measure the numbers that make theories work, and that the theories themselves, are robust and unambiguous such that the resultant models can be used to assist the poultry Industry to become more efficient especially when designing feeds and feeding programmes.