Optimizing sheep growth curves using a meta-heuristic algorithm

Area of Science:

• Animal Science
• Quantitative Genetics
• Bioinformatics

Background:

• Sheep are vital livestock for meat, milk, and wool.
• Accurate growth modeling is crucial for efficient animal production and genetic selection.

Purpose of the Study:

• To evaluate growth curve models for crossbred lambs.
• To determine optimal parameter values for growth models using a hybrid metaheuristic algorithm.
• To assess the goodness of fit and accuracy of simulated growth curves.

Main Methods:

• Utilized a hybrid metaheuristic algorithm (SAGAC), combining simulated annealing (SA) and genetic algorithm (GA).
• Employed observed age and weight data from crossbred lambs.
• Assessed model fit using R², MAE, RMSE, and MAPE.

Main Results:

• Simulated 30 crossbreed combinations, analyzing average weight and growth curves.
• Identified specific crossbreeding scenarios with promising growth patterns.
• The SAGAC algorithm effectively aligned simulated with observed growth data.

Conclusions:

• The SAGAC simulation approach aids geneticists in predicting crossbreeding outcomes.
• This method offers a valuable tool for saving time and financial resources in sheep breeding research.
• Accurate growth curve modeling can optimize livestock management and genetic improvement strategies.