Phenomic Selection for Hybrid Rapeseed Breeding

Area of Science:

• Agricultural Science
• Plant Breeding
• Spectroscopy

Background:

• Genomic selection (GS) is a standard tool in plant breeding.
• Phenomic selection (PS) offers a low-cost, high-throughput alternative using spectral data.
• The efficacy of PS for predicting traits in hybrid generations using parental data is largely unexplored.

Purpose of the Study:

• To evaluate the potential of phenomic selection for hybrid rapeseed breeding.
• To assess if spectral data from parental genotypes can predict traits in hybrid offspring.
• To compare PS with GS in the context of hybrid breeding.

Main Methods:

• Near-infrared spectroscopy (NIRS) data were collected from parental lines and 410 test hybrids.
• Spectral data were used with linear and nonlinear models to predict traits like seed yield and plant height.
• Predictions were made within the hybrid generation and across generations (parental to hybrid).

Main Results:

• Phenomic selection within the hybrid generation surpassed genomic selection for seed yield and plant height.
• PS performance was robust even with spectral data from single locations and less influenced by population structure.
• Phenomic prediction across generations proved feasible, with parental spectral data offering competitive prediction accuracy against GS.

Conclusions:

• Phenomic selection is a viable and promising strategy for hybrid rapeseed breeding.
• Utilizing spectral data from parental lines for phenomic prediction in hybrids is effective and competitive with genomic selection.
• PS, leveraging routinely collected NIRS data, can be seamlessly integrated into rapeseed breeding programs without additional costs.