Evaluating of the efficiency of AMMI and BLUP models and their integration for identifying high-yielding durum wheat (Triticum turgidum L. var. durum) genotypes adapted to warm rainfed regions of Iran

The aim of this study was to evaluate the efficiency of yield stability analysis models and to assess genotype × environment interaction effect on grain yield of 20 durum wheat genotypes for identifying high yielding and adapted genotypes by BLU and AMMI models using experimental data of four cropping cycles (2009-2013) in five filed stations in warm rainfed regions of Iran. The results of Likelihood ratio test (LRT) showed that the effect of genotype and genotype × environment interaction on grain yield was significant. Therefore, the best linear unbiased predictors (BLUPs) analysis was considered appropriate for these data. According to AMMI stability value (ASV) index, genotypes 11, 8, 14, 19 and 16 had more yield stability. Simultaneous selection index (ssiASV) based on ASV identified genotypes 8, 11, 4 and 10 in terms of grain yield and yield stability as superior genotypes. Given that by using these simultaneous selection indices, genotypes with different patterns for multivariate trials can be considered similar, the results can be misleading. Based on the first two main components, AMMI2 biplot diagram identified genotypes 19, 3, 14 and 11 as genotypes with yield stability. The results of the mosaic diagram showed that the contribution of genotype and genotype × environment interaction were 14.94% and 85.06% of the total variation, respectively. Based on weighted average of absolute scores (WAASBY) index using BLUP analysis, genotypes 4, 8, 11, 10 and 7 were identified as high yielding with yield stability. In general, usin mixed model as well as all the components in calculating the WAASBY index, it can be concluded that this index is superior to other indices.