抄録
With the increasing demand for high-quality raw materials in the food processing and agricultural industries, optimizing potato starch quality has become a key factor in enhancing product value and meeting diverse market needs. This study systematically explores methods for improving potato starch quality through the integration of big data and phenotypic analysis. Focusing on key phenotypic traits such as starch granule size, gelatinization characteristics, viscosity, and gel strength, the research integrates multidimensional data from genomics, transcriptomics, and environmental factors to reveal the complex regulatory mechanisms underlying genotype–environment interactions in starch quality formation. By applying big data analytics and machine learning algorithms, a multidimensional predictive model for starch quality was developed, enabling effective fusion of high-throughput phenotypic and genomic data to support data-driven decision-making in targeted breeding. Moreover, the application of artificial intelligence (AI) in phenotypic feature recognition and model optimization significantly improved analytical efficiency and predictive accuracy. Practical case studies demonstrated the potential of big data and AI technologies in starch quality improvement, and future research directions were discussed, including data standardization, cross-environment adaptability modeling, and the development of intelligent decision-support systems. This study aims to establish a data-driven, systematic research framework to provide theoretical and technical support for precise phenotypic analysis and molecular breeding of potato starch quality, thereby promoting agricultural intelligence and the sustainable development of the food industry.
