Smart Breeding: Integrating AI, Genomics and Phenomics for Next-Generation Crops: A Review

Abstract

The convergence of artificial intelligence (AI), genomics and phenomics is ushering in a new era of smart breeding a paradigm that promises to dramatically accelerate genetic gain while reducing the time and cost associated with developing elite crop varieties. Conventional plant breeding, though enormously successful over the past century, is increasingly challenged by a rapidly changing climate, a growing global population projected to reach nearly 10 billion by 2050 and the biological complexity of quantitative traits. Smart breeding leverages exponential growth in genomic data, high-throughput phenotyping platforms and the analytical power of machine learning and deep learning algorithms to navigate these challenges. This review synthesizes the current state of knowledge across three interdependent pillars: AI and machine learning for genomic selection, trait prediction and decision support; next-generation sequencing and multi-omics tools that have transformed our understanding of crop genetic architecture; and field and controlled-environment phenomics platforms that bridge the genotype phenotype gap. Further discuss integration through digital twins, knowledge graphs and federated learning frameworks and examine applications in gene editing, stress tolerance and yield improvement. Key challenges data standardization, interpretability of black-box models, regulatory frameworks and equitable access are critically assessed and a roadmap for the next decade of smart breeding is proposed. Another point highlighted in this review is the need to conduct collaborative and interdisciplinary research to achieve the full potential of smart breeding technologies. It emphasizes the necessity of capacity-building, data sharing systems and policy support to provide sustainable and inclusive agricultural growth. Moreover, the paper highlights the importance of new innovations in developing resilient and productive and future-oriented crop systems.

Plant phenotyping relevance

AI・ゲノミクス・フェノミクスを統合するレビューであり、高スループット表現型解析プラットフォーム、形質予測、機械学習による表現型解析を主要な構成要素として扱っているため、方法レビューとして採択。

Code and data availability

公開論文であることは確認できましたが、現在の公式API・許可済み取得経路では本文を自動取得できませんでした。

No evidence-backed public reproduction asset is currently recorded.