Machine Learning in Legume Breeding: Modeling Genotype and Environment Interactions

In this review, addresses the importance of machine learning techniques in modelling genotype-environment (G×E) interactions in legume breeding. Agricultural production is greatly affected by climate change and environmental stressors, and a better understanding of these interactions is critical for the development of environmentally adaptive and high-yielding varieties. Genotype-environmental interactions are valuable for understanding how genetic and environmental factors affect plant performance. The study reported that machine learning algorithms such as support vector machines (SVM), random forests (RF), deep learning (DL) and neural networks (ANN) are used for processing high-dimensional data sets, assessing genotype stability and modelling response to environmental stresses. These techniques offer significant advantages in the evaluation of complex traits such as yield, quality and adaptability in legumes. Machine learning approaches overcome the shortcomings of traditional methods and provide more powerful tools, especially in processing large data sets, modelling environmental variables and predicting genetic performance. The effective use of these methods has great potential to increase sustainability in agricultural production and to develop varieties that are resilient to climate change