Articles

The growing demand for resilient crops, climate change and the availability of high-throughput sequencing and sensor technologies have created an unprecedented opportunity for data-driven crop improvement. Multi-omics technologies (genomics, transcriptomics, proteomics, metabolomics, epigenomics and phenomics) capture different layers of biological information, yet analysing each layer separately loses the holistic view of how these molecules collectively shape phenotype. Integrating these heterogeneous datasets with artificial intelligence (AI) can reveal complex gene-environment interactions and accelerate trait improvement. This review, written from a plant breeder’s perspective, summarizes the current state of AI-assisted multi-omics integration in plant genetics. We describe the omics landscape, discuss machine-learning algorithms and integrative frameworks, review applications in breeding (stress tolerance, disease resistance, yield and quality traits), and examine challenges such as data heterogeneity, model interpretability and equitable data sharing. Finally, we offer recommendations for the next generation of AI-enabled plant breeding programs.