Updating the Cornell Net Carbohydrate and Protein System feed library and analyzing model sensitivity to feed inputs

Area of Science:

• Animal Nutrition
• Nutritional Modeling
• Agricultural Science

Background:

• The Cornell Net Carbohydrate and Protein System (CNCPS) is a widely used nutritional model for animal production.
• Accurate ingredient composition data is crucial for effective diet formulation.
• Previous versions may have had limitations in predicting nutrient supply.

Purpose of the Study:

• To update and refine the CNCPS feed library with contemporary data.
• To improve the prediction accuracy of dietary energy and protein supply.
• To enhance the model's biological description and user-friendliness.

Main Methods:

• Evaluation of approximately 800 feed ingredients against commercial laboratory data.
• Development of a multistep approach (linear regression, matrix regression, optimization) for predicting uncertain values.
• Reclassification of the protein A fraction to ammonia and updating amino acid profiles.

Main Results:

• The updated CNCPS feed library provides more precise predictions of dietary energy and protein.
• Model sensitivity analyses indicated metabolizable energy prediction is sensitive to feed chemistry, and metabolizable protein to digestion rates.
• The reclassification of protein A and updated amino acid profiles improve predictions of metabolizable protein.

Conclusions:

• Updates to the CNCPS feed library enhance its utility for formulating animal rations.
• The refined model offers a more accurate description of nutrient supply and animal requirements.
• While on-farm analysis is recommended, the updated library serves as a robust platform for ration formulation.