Machine learning in plant science and plant breeding.

Summary

This narrative review surveys the application of machine learning methods in plant science and breeding, addressing the challenge of extracting meaning from large, complex datasets generated by modern phenotyping and genotyping technologies. The authors examine machine learning approaches across multiple phenotypic scales—from biochemical to agronomic traits—and their utility in connecting genotypic data to phenotypic outcomes, illustrating the expanding toolkit machine learning provides for plant research.

Regional applicability

As a methodology review without geographic specificity, the findings are globally applicable to plant breeding and phenotyping research. UK plant breeding programmes and agricultural research institutions would benefit from considering these machine learning approaches in their own data integration and trait prediction workflows.

Key measures

Applications of machine learning across phenotype levels (biochemical, morphological, yield); genotype–phenotype associations

Outcomes reported

The review examines machine learning methods applied across multiple phenotype levels in plant science, from biochemical to yield traits, and their role in connecting genotypes to phenotypes. It synthesises how machine learning enables extraction of meaningful patterns from large, complex plant datasets.

Topic tags