夏南牛基因组选择需求及机器学习在牛育种中的应用进展OA

Xianan cattle,a premium beef breed independently developed in China,combine the high produc-tivity of introduced breeds with the adaptability of local yellow cattle.However,their complex genetic back-ground(hybrid origin,moderate diversity,and small population size)limits the accuracy of traditional ge-nomic selection methods.This review systematically outlines the specific requirements for genomic selec-tion breeding in Xianan cattle and summarizes recent advances in genomic selection and machine learning algorithms within the cattle breeding sector.It focuses on the application of machine learning approaches,such as regularized regression,support vector machines,and neural networks in improving the accuracy of breeding value prediction for complex quantitative traits,multiple-trait modeling,and gene-environment in-teractions.Furthermore,the paper discusses the potential of transfer learning and explainable artificial in-telligence(AI)in advancing Xianan cattle breeding,offering new perspectives for establishing a precise and intelligent genomic breeding system.