Machine learning bridges omics sciences and plant breeding.

Area of Science:

• Plant Science
• Bioinformatics
• Computational Biology

Background:

• Bridging fundamental biological research and applied plant breeding is crucial for crop improvement.
• Omics data and biological knowledge offer vast potential for enhancing plant breeding strategies.

Purpose of the Study:

• To review the applications of machine learning (ML) in analyzing multi-omics data for plant breeding.
• To highlight how ML can translate biological insights into precision-designed breeding programs.
• To discuss the role of ML in advancing knowledge-driven and data-driven plant breeding.

Main Methods:

• Review of machine learning techniques applied to plant multi-omics data.
• Analysis of ML for data dimensionality reduction, gene network inference, and gene prioritization.
• Examination of deep learning in plant phenomics and ML in genomic selection.

Main Results:

• ML facilitates the interpretation of complex biological knowledge and omics data in plants.
• Applications include inferring gene regulatory networks and discovering/prioritizing target genes.
• ML models correlations between genotypes, phenotypes, and environments for genomic selection.

Conclusions:

• Machine learning is a powerful tool for integrating biological knowledge and omics data in plant breeding.
• ML enables a deeper understanding of trait regulation and identification of key genes for molecular design breeding.
• ML advances genomic selection and facilitates data-driven genomic design breeding for improved crop traits.