Genome wide selection (GS) was proposed by meuwissen et al in 2001. It refers to the prediction of quantitative traits that depend on genotype, have great environmental impact and are controlled by micro genes by using high-density molecular markers covering the whole genome.
As the cost of SNP typing is lower than that of phenotype typing, genome-wide selection has become the best scheme to solve the selection of quantitative traits greatly affected by the environment and controlled by micro effect genes. GS is the most advanced core technology of molecular breeding at present. Bayer Monsanto, kedihua and other large international companies have integrated GS into commercial seed breeding.
When the SNP marker density is high enough, all quantitative trait loci (QTLs) controlling the trait are unbalanced or associated with at least one SNP marker.
Select representative strains or individuals, use high-density SNP markers covering the whole genome for typing and phenotypic identification, and then construct and verify the prediction model. The same group of high-density SNP markers covering the whole genome were used to type new high-generation or stable breeding materials, and then the verified prediction model was used to calculate the genome estimated breeding value, and a few lines with high breeding value were selected for field phenotypic evaluation.
Application of GS prediction model
Create a personalized genome-wide selection prediction model according to specific germplasm resources or breeding projects, including:
- Predict the Heterosis of combinations and screen strong and excellent hybrid combinations.
- Choose excellent parents.
- The model was used to design hybrid combinations for constructing isolated populations.