MsChick: A dynamic mechanistic model for simulating nutrient partitioning and growth in broilers

Mechanistic models are increasingly required in poultry nutrition to account for genetic progress, refine nutrient supply, and improve economic and environmental sustainability. This study presents MsChick (Model simulation for Chicken), a dynamic mechanistic model that simulates daily feed intake, growth, body composition, and selected carcass traits in broilers using a net energy (NE) and standardized ileal digestible amino acid (SID-AA) framework. Energy is partitioned first to maintenance and thermoregulation/activity, then to protein deposition, with remaining energy retained as lipid. Digestible amino acids are allocated to feathers, maintenance-related losses, and feather-free body protein deposition. A strain × sex phenotype is represented by four phenotypic parameters: mean potential feather-free body-protein deposition ( meanBPD ), precocity of feather-free body-protein deposition ( precBPD ), and estimated NE intake at 1 and 2 kg body weight ( NEI1 and NEI2 , respectively). Ad libitum intake is simulated from a body weight-dependent NE intake curve expressed relative to maintenance, whereas daily feed intake can be supplied directly; both modes enter the same nutrient-partitioning core. The model was calibrated to breeder performance objectives for Ross 308 and Cobb 500 males and females and evaluated against independent datasets including body-composition measurements and individual performance records. Following calibration, MsChick closely reproduced breeder body weight and daily feed-intake trajectories. In independent evaluation, agreement was strongest for body weight and body water, whereas lipid, feather weight, feather-free body protein, and ash showed greater uncertainty. Sensitivity analysis identified meanBPD as the principal lean-growth parameter, whereas FHPinit , the baseline fasting heat-production coefficient, influenced late-stage lipid deposition. MsChick provides an explicitly documented average-bird mechanistic framework in which phenotype, diet composition, and simulated or directly supplied intake can be interpreted through a common nutrient-partitioning structure. In the current version, internal partitioning variables should be interpreted as literature-informed, model-derived diagnostic indicators rather than direct physiological measurements.