Leveraging AI and integrated genomic–enviromic prediction for intelligent sugarcane breeding

Traditional sugarcane breeding, reliant on phenotypic selection, is being transformed by genomic tools. However, the crop's highly polyploid genome and significant genotype-by-environment interactions pose challenges that conventional models cannot adequately address. Although the integrated genomic-enviromic prediction (iGEP) framework offers a promising path forward, its application to a complex clonal crop such as sugarcane requires significant extension. This review provides the first comprehensive road map for implementing iGEP in sugarcane, systematically addressing its unique biological constraints, and synthesizes a tailored "three-model" computational framework (genetic, environmental, and phenotypic) to decode polyploid allelic dosage, quantify high-resolution environmental drivers through an "isoenvironment" design, and predict clonal performance. In addition, we describe extensions of artificial intelligence (AI) and iGEP models to leverage clonal propagation, optimize multi-trait selection, and overcome perennial ratoon dynamics. Finally, we present a phased road map for construction of an AI model, outlining a transformative path from digitization to synthetic design. By combining cutting-edge predictive analytics with the distinctive biology of sugarcane, this work establishes a new paradigm for accelerating genetic gain in this vital crop and offers a transferable strategy for other species with complex genomes.