CBGG

Late blight is a disease responsible for major losses to the potato industry and breeding new resistant varieties is a sustainable way to fight late blight and reduce the need for chemical treatments. This study aimed at understanding the genetic basis of late blight resistance in a commercial population of potatoes. Historical records from a large breeding program were used. Different statistical models were evaluated to determine the contribution of additive genetic variance, non-additive genetic variance, and genotype-by-decade of testing interaction variance to the total phenotypic variance. GWAS were conducted to identify genomic regions associated with late blight resistance and the part of the additive genetic variance explained by these regions were estimated. Breeding values for resistance toward late blight were predicted using four different modeling approaches. In the two last models the most significant SNP of significant genomic regions were included as fixed effects. Predictive ability was assessed using five-fold cross-validation by correlating corrected phenotypes or breeding values predicted using all data with those predicted without the phenotypes in the validation populations. Results showed that late blight resistance was highly heritable and that the contribution of the genotype-by-decade interaction was significant, indicating differential resistance toward different late blight strains dominating in each decade. Three genomic regions significantly affecting late blight resistance explained only 13% of the total additive genetic variance in resistance. Genomic breeding values (GEBV) showed high predictive ability suggesting that genetic or genomic selection could be an effective tool to develop new late blight resistant varieties. A strategy combining genomic selection and including knowledge on genomic regions associated with resistance is advantageous.

Phytophthora infestans; tetraploid potatoes; resistance genes; host pathogen coevolution; genomic selection