PLANT CELL BIOTECHNOLOGY AND MOLECULAR BIOLOGY

The interaction between genotype and environment (G×E) reduces the heritability estimate and thus the efficiency of selection. Among many methods suggested for assessment of G×E interaction, AMMI analysis and GGE-biplot model are the best in giving detailed information. The aims of the present investigation were: (i) to estimate the G×E interaction for 37 quinoa genotypes and (ii) to identify the stable and/or adaptable genotypes across 12 environments (two fertilizer types × three N rates × two seasons) by using the AMMI and GGE-biplot models. In each environment, the randomized complete blocks design with three replications was used. Analysis of variance of   AMMI indicated that genotype (G), environment (E) and G×E interaction had a highly significant influence on quinoa seed yield/ha. Based on AMMI model, NAD-1-W (G2), and Agritec-Beige (G6) could be considered stable across the test environments and among the five highest seed yielding genotypes in this experiment. R-103 (G9), Sekam-Bitter (G4), Nat-1(G8), ICBA-Q5 (G14) and CICA-17 (G12) were the highest yielding genotypes, but were considered the most unstable genotypes. The optimum environment E9 (organic fertilizer, low-N, 2018/19 season) is the most stable based on IPCA_e-1 and ASV scores; therefore it is the lowest interactive environment for seed yield/ha, and is considered the ideal one for selection of adapted genotypes under organic fertilizer conditions. The environments E1 (organic fertilizer, high-N, 2017/18 season) showed good discrimination ability and representativeness, making it ideal for evaluating the genotypes of quinoa. Based on GGE-biplot model, Nat-1 (G8) was the winning genotype for the 1^st mega environment that consisted of two environments. Sekam-Bitter (G4) was the winning genotype for the 2^nd mega environment that consisted of four environments. R-103 (G9) was the winning genotype for the 3^rd mega environment that consisted of six environments.