Analyzing Multi-Environment DUS Test Data for Stability Assessment and Cross-Environment Characteristic Association in Maize Breeding

Maize is a staple crop supporting global food security. Comprehensive evaluation of its phenotypic variation and environmental adaptability is urgently needed. In this study, three mutually independent maize germplasm panels with a total of 1197 hybrids were characterized using 39 standardized DUS traits: Panel A (783 hybrids, single location, 2023), Panel B (129 hybrids, two locations in the same year), and Panel C (285 hybrids, single location across two consecutive years). Tassel- and anthocyanin-related traits exhibited wide phenotypic variation and strong germplasm discrimination ability, while leaf margin anthocyanin (Char. 8) displayed rare genotypes with unique identification values. Hierarchical clustering divided all materials into eight distinct groups. PCA revealed weak overall phenotypic differentiation, which is likely driven by long-term directional selection in modern maize breeding. AMMI-based stability evaluation demonstrated that grain morphological traits (Char. 33, Char. 35) maintained stable performance across locations and years, whereas anthocyanin-related descriptors (Char. 8, Char. 14) showed severe environmental plasticity. Using the classic Multi-Trait Stability Index (MTSI), we screened top stable genotypes: 77, 69 and 52 for cross-location trials, and N90, N85 and N107 for multi-year trials. BLUP five-fold cross-validation revealed that grain morphological traits achieved high cross-location predictive correlation (r2 > 0.80), whereas cross-year predictability was notably lower, reflecting greater influence of annual climatic fluctuations. Collectively, this study constructs a standardized analytical pipeline to mine routine multi-environment DUS records, identifies stable elite germplasm, and provides data-driven references for optimizing maize breeding and improving official DUS testing efficiency.