Pan Genome to Pan Phenome provides a mechanistic framework for translating hidden genetic diversity into predictive wheat breeding

Wheat breeding has entered the genomic era, yet most breeding decisions still rely primarily on sequence polymorphisms, leaving a substantial proportion of functional genomic diversity unexplored. Hidden variation, including presence–absence variation, structural variants, copy-number variation, transposable element insertions, and regulatory polymorphisms, constitutes a major source of phenotypic diversity but remains insufficiently integrated into breeding pipelines. The principal challenge is no longer identifying additional genetic variants but understanding how this hidden genomic diversity is translated into complex phenotypes through successive biological processes. Here, we propose the Pan-Genome to Pan-Phenome (PG2P) framework, a conceptual and mechanistic framework synthesized from current knowledge that connects hidden genomic variation with regulatory networks, epigenetic regulation, protein interactions, metabolic reprogramming, developmental dynamics, and ultimately the wheat pan-phenome. We further discuss how this biological continuum facilitates causal variant discovery, improves trait prediction, and supports predictive precision breeding by integrating functional genomic information with genomic prediction and selection. Finally, we discuss key challenges and future opportunities, including graph pan-genomes, single-cell and spatial omics, pangenome-aware prediction, digital phenotyping, genome foundation models, and digital twins. The PG2P framework provides a conceptual roadmap for translating hidden genomic diversity into predictive biological knowledge for next-generation wheat improvement.