Predicting complex phenotypes using multi-omics data in maize.

Area of Science:

• Plant genomics and systems biology
• Agricultural biotechnology and breeding

Background:

• Predicting complex plant traits is challenging due to intricate genetic, regulatory, and environmental interactions.
• Accurate trait prediction and identification of genetic elements are crucial for plant breeding, systems biology, and biotechnology.

Purpose of the Study:

• To evaluate if multi-omic datasets (genomic, transcriptomic, phenomic) enhance predictive accuracy for diverse maize phenotypes across multiple environments.
• To compare the performance of linear (rrBLUP) and nonlinear (support vector regression) models using single- and multi-omics inputs.

Main Methods:

• Utilized genomic markers, field-based transcriptomic data, and drone-derived phenomic data for 129 maize phenotypes across nine environments.
• Trained and compared linear and nonlinear predictive models using single-omics and integrated multi-omics datasets.

Main Results:

• Multi-omics models consistently outperformed single-omics models, with genomic and transcriptomic data providing distinct biological insights.
• Phenomic data alone showed lower predictive power but improved predictions for specific traits like root architecture.
• Transcriptomic data facilitated accurate cross-environment trait prediction and captured genotype-by-environment (G×E) interactions.

Conclusions:

• Integrating transcriptomic and phenomic data with genotypes significantly enhances maize trait prediction and model generalizability across environments.
• This multi-omics approach provides deeper insights into the genetic and regulatory architecture of agriculturally important plant traits.