The genotype-by-environment (G \(\times\) E) interaction represents the differential response of genotypes to the various environments in which they are grown; ignoring this interaction during the selection process can lead to suboptimal outcomes. This study aimed to identify superior soybean genotypes across multiple environments and determine key environmental features that causes the effect of the G \(\times\) E interaction. Using a dataset of 104 genotypes evaluated across 16 environments, we performed spatial analyses using splines in the first step, followed by factor analytic (FA) models in the second step. FA selection tools were employed to select genotypes with good overall performance, broad stability, and specific adaptation to different environments. We investigated the influence of 29 environmental features on G \(\times\) E interaction using correlations with the FA factor loadings. Fifteen genotypes with good general adaptation were recommended, and the genotype with the highest performance was selected for specific adaptation. The features identified as most influential in the G \(\times\) E interaction included sand content, total radiation, net radiation, nitrogen content, silt content, downward longwave radiation, dissolved organic carbon content, and cation exchange capacity. The methodologies presented in this study provide valuable insights for breeding programs by enhancing the understanding of specific environmental features and genotype responses through their interactions.
