Genomic Selection in Preliminary Yield Trials in a Winter Wheat Breeding Program.

Area of Science:

• Plant breeding
• Quantitative genetics
• Genomics

Background:

• Genomic prediction (GP) is increasingly used in crop science to forecast unobserved plant traits.
• Utilizing predicted phenotypes for selection decisions is a key research focus in modern agriculture.

Purpose of the Study:

• To evaluate GP for predicting grain yield in crop plants.
• To compare the effectiveness of genomic selection against traditional phenotypic selection.
• To assess the integration of GP into breeding programs for autogamous crops.

Main Methods:

• Analysis of four independent nurseries (F3:6 and F3:7 lines) across multiple locations and years.
• Yield data analyzed using mixed models accounting for experimental design and spatial variation.
• Genotyping using Genotype-by-sequencing (GBS) to obtain ~27,000 high-quality SNPs.

Main Results:

• Average genomic predictive ability ranged from 0.23 to 0.55 within years and 0.17 to 0.28 across years.
• Lines with both high genomic estimated breeding values (GEBV) and phenotypic values (BLUP) were advanced more consistently.
• Advancement selection efficiency increased significantly when 50% of lines from the testing year were included in the training set.

Conclusions:

• Genomic prediction is a viable tool for enhancing grain yield prediction in crop breeding.
• Combining genomic and phenotypic data improves the accuracy and efficiency of selecting superior lines.
• The findings suggest that evaluating only 50% of lines annually may be feasible, optimizing breeding program resources.