抄録
Calcium, as an essential macronutrient for plants, plays an irreplaceable role in maintaining cell wall stability, mediating calcium signaling, regulating growth and development, and enhancing stress adaptation; its uptake and utilization efficiency directly influence crop yield formation and quality safety, making it a key scientific issue in nutrient-efficient utilization and genetic improvement. Focusing on improving crop calcium uptake and use efficiency, this study systematically reviews the physiological functions of calcium in plants, its transmembrane transport pathways, and its signaling regulatory mechanisms, and further introduces artificial intelligence (AI) to construct a new analytical framework for genotype × environment (G×E) interactions. Methodologically, this review evaluates the applications of machine learning, deep learning, and big data analytics in multi-environment phenotyping, calcium-use efficiency prediction, and elite genotype identification, with emphasis on integrative strategies that combine multi-omics datasets (genomic, transcriptomic, phenomic, and environmental data) with advanced AI models. The findings indicate that AI-enabled G×E modeling can effectively identify key regulatory factors underlying calcium uptake and utilization, enhance the accuracy of predicting high-efficiency calcium-use genotypes under complex environments, and provide technical support for precision breeding of calcium-efficient and stress-resilient crop varieties. Overall, the study concludes that the deep integration of AI and calcium nutrition biology holds strong potential to overcome bottlenecks in conventional breeding related to trait dissection and selection efficiency, offering new theoretical foundations and methodological pathways for intelligent, targeted molecular breeding aimed at improving calcium-use efficiency, while simultaneously posing new challenges regarding data quality, model interpretability, and cross-environment generalization.
