Genomic Selection in Red Clover: Issues and Solutions

Genomic selection (GS) has been expected to be an effective approach for marker assisted selection (MAS) in forage species, because it is based on the idea of recurrent selection, which has been traditionally employed in forage breeding programs. However, the operation cost of GS is still expensive due to necessity of genome wide genotyping data of both training and breeding populations.

For reducing the number of used markers in GS without losing accuracy of selection, we have developed MAGIC (Multi-parent Advanced Generation Inter-Cross) population in red clover. Each 200 plants were investigated morphological traits in Japan, India, Switzerland and Russia for two years. A total of 384 SNP and 75 SSR markers were used for genotyping of the investigated plants. In parallel, we have developed an advanced GS algorithm named EGGS (Ensemble-based Genetic and Genomic Selection), which chooses a set of markers for investigation of Breeding score. By using the EGGS method, we select each twenty markers for selection of flowering date investigated in the four regions. Of the each twenty markers, several markers were common in the four regions, while the others were not. The result suggested that EGGS is an effect method for selection of quantitative traits in forage.