BLUP (Best Linear Unbiased Predictors) analysis for the selection of superior yellow diploid potato genotypes (Solanum tuberosum group Phureja)

Abstract

One of the major challenges that breeders face is the differential response of genotypes from one environment to another, known as the genotype × environmental interaction (GxE). The optimal procedure with the restricted maximum likelihood/best linear unbiased predictor (REML/BLUP) allows simultaneous estimation of genetic parameters and prediction of genotypic values. BLUP predictors are an alternative to the narrowing of biased values, which are based on variances of genotype to determine the response value, as a complement to the selection index (SI). The ESIM (Eigenvalue Selection Index) selects genotypes based on two or more variables or selection characters as long as the economic matrix possesses the appropriate values for highlighting the desired response variable. Three stages of selection were evaluated in an advanced diploid potato improvement program. BLUP values were obtained for the yield and specific gravity variables, used to determine the genetic parameters and the SI. The genetic gain for yield corresponded to 1.228 kg/plant with a heritability (H2)=0.82, while the GA for GE was 0.02 with an H2=0.935. The SI from the BLUP values selected in the final stages of the three new cultivars (Criolla Dorada, Criolla Ocarina and Criolla Sua Pa) was registered at the Instituto Colombiano Agropecuario (ICA). Although BLUE and BLUP are highly correlated, the BLUP/ESIM analysis has an advantage as a predictor because it reduces responses to the environmental effect, efficiently selecting genotypes with a high agronomic potential.