Abstract
Climate change–driven drought, heat, salinity, and combined stresses pose major challenges to crop productivity and food security. Multi-omics technologies, high-throughput phenotyping, and artificial intelligence (AI) have emerged as powerful tools for elucidating stress-response mechanisms and accelerating crop resilience breeding. This review synthesizes recent advances in genomics, transcriptomics, proteomics, metabolomics, phenomics, and AI-driven modeling, and critically evaluates their roles in decoding genotype–phenotype–environment interactions and predicting complex adaptive traits. Evidence suggests that the convergence of multi-omics, phenomics, and AI substantially enhances the precision, scalability, and predictive power of resilience-oriented breeding. Nevertheless, challenges related to data heterogeneity, standardization, model interpretability, causal inference, and field-level implementation continue to constrain practical deployment. Future research should prioritize causal and explainable AI, multi-scale data integration, digital phenotyping, and closed-loop breeding systems linking gene discovery with breeding decisions. This review highlights the emerging convergence of multi-omics and AI as a transformative paradigm for climate-resilient crop breeding and provides a strategic framework for developing next-generation intelligent breeding systems.
