The potato tuber is known as a rich source of essential nutrients, used throughout the world. Although potatobreeding programs share some priorities, the major objective is to increase the genetic potential for yield through breeding or to eliminate hazards that reduce yield. Glycoalkaloids, which are considered a serious hazard to human health, accumulate naturally in potatoes during growth, harvesting, transportation, and storage. Here, we used the AMMI (additive main effects and multiplicative interaction) and GGE (Genotype main effect and genotype by environment interaction) biplot model, to evaluate tuber yield stability and glycoalkaloid content in six potato cultivars across three locations during 2012/2013. The environment on tuber yield had the greatest effect and accounted for 33.0% of the total sum squares; genotypes accounted for 3.8% and G×E interaction accounted for 11.1% which is the nest highest contribution. Conversely, the genotype on glycoalkaloid had the greatest effect and accounted for 82.4% of the total sum squares), whereas environment and G×E effects on this trait accounted for only 0.4% and 3.7%, respectively. Furthermore, potato genotype ‘Superior’, which covers most of the cultivated area, exhibited high yield performance with stability. ‘Goun’, which showed lower glycoalkaloid content, was the most suitable and desirable genotype. Results showed that, while tuber yield was more affected by the environment, glycoalkaloid content was more dependent on genotype. Further, the use of the AMMI and GGE biplot model generated more interactive visuals, facilitated the identification of superior genotypes, and suggested decisions on a variety of recommendations for specific environments.

Application of genotype by environment (G~;~times~;E ) interaction would be used for identifying optimum test condition of the varietal adaptation in the establishment of breeding purpose. Yield and yield components were used to perform additive main effect and multiplicative interaction (AMMI) analysis. Significant difference for G~;~times~;E interaction were observed for all variable examined. For yield, 0.18 of total sum of squares corresponded to G~;~times~;E interaction. Correlation analysis was carried out between genotypic scores of the first interaction principal component axis (IPCA 1) for agronomic characters. Significant correlations were observed between IPCA 1 for yield and capsule bearing stem length (CBSL), number of capsule per plant (NOC). The biplot of grain yield means for IPCA1 which accounted for 34% of the variation in total treatment sums of squares showed different reaction according to G~;~times~;E interaction, genotypes and environments. Taegu showed relatively lower positive IPCA1 scores, and it also showed smaller coefficient variation of yield mean where it is recommendable as a optimal site for the sesame cultivar adaptation and evaluation trial. In case of variables, Yangbaek and M1 showed relatively lower IPCA1 scores, but the score direction showed opposite each other on the graph. Ansan, Miryang1, Miryang4, and Miryang6 seemed to be similar group in view of yield response against IPCA1 scores. These results will be helpful to select experimental site for sesame in Korea to minimize G~;~times~;E interaction for the selection of promising genotype with higher stability.