APPLYING ARTIFICIAL INTELLIGENCE IN MUTATION BREEDING

ABSTRACT

Many studies have highlighted the transformative role of artificial intelligence (AI) and machine learning (ML) in the field of agriculture. However, works on the application of AI algorithms in breeding initiatives for improved varieties remain underexplored. To bridge this gap, this study presents a computational modelling approach to cowpea mutation breeding. Three cowpea landraces were induced with chemical mutagen (ethyl methane sulphonate - EMS) and physical mutagen (gamma irradiation) to enhance their growth and yield. The dosages used included gamma ray treatment doses 200Gy, 400Gy, 600Gy and 800Gy, along with a fixed EMS concentration of 0.372 were applied. Genetic algorithm (GA) was used as an AI tool to model the mutation induction in search of candidate solution desired yield and drought-tolerant traits. The AI model projected that the optimal cowpea solution could be obtained in the 9412^th generation. Such projection is useful for informed decision-making in breeding programmes and offering a novel tool for researchers and breeders aiming to develop climate-resilient and high-yielding crop varieties.

Keywords: Artificial intelligence, Genetic algorithm, Mutation breeding, Optimization, Predictive analytics Python programming