Click Breeding for programmable crop design
Genomic prediction alone increasingly struggles to generalize across environments for complex polygenic traits shaped by nonadditive effects and G × E interactions.
AI-driven design agents can generate and evaluate multigenerational breeding strategies under explicit objectives, constraints, and uncertainty—moving beyond candidate ranking.
Trait-level digital twins integrating genetics, environment, and physiology enable in silico design-before-build and riskaware decision-making with calibrated uncertainty.
Governance-by-design, embedded in step-by-step experimental workflows, enables traceability and biosafety compliance as a built-in feature of AIenabled breeding pipelines.
https://www.cell.com/trends/plant-science/fulltext/S1360-1385(26)00210-4
