Selection of adaptable lablab ( Lablab purpureus L.) genotypes through dry matter yield performance and multi‐trait stability analysis

Lablab ( Lablab purpureus L.) is a nutritious, multipurpose legume used globally as livestock fodder and human food. A study in Northern Ethiopia (2021–2023) evaluated genotype‐by‐environment interaction (GEI) in multi‐environment trials to identify stable lablab genotypes based on dry matter yield (DMY) performance and a multi‐trait stability index (MTSI). Five lablab genotypes were evaluated across three locations over three consecutive years using a randomized complete block design with three replications. The measured traits included days to emergence (DE), days to 50% flowering (DF), number of branches (BRN), plant height (PH), and DMY. Combined analysis of variance (ANOVA), additive main effects and multiplicative interaction (AMMI) ANOVA, and Eberhart and Russell regression were analyzed. Genotype plus genotype‐by‐environment (GGE) bi‐plots, AMMI1 and AMMI2 bi‐plots, principal component analysis, and correlation network plots for lablab traits were created. The MTSI of the agronomic traits of lablab genotypes was analyzed. AMMI‐based, best linear unbiased prediction (BLUP)‐based, parametric, and nonparametric stability statistics for the DMY of lablab genotypes were computed using R software. AMMI ANOVA revealed significant ( p < 0.001) effects of environment (E) (50.94% sum of squares [SS]), genotype (G) (7.15% SS), and GEI (14.52% SS) on the DMY of lablab genotypes. The likelihood ratio test (LRT) revealed a significant genotype effect for DE, DF, BRN, PH, and DMY. Likelihood ratio tests of the genotype‐by‐environment interaction (LRTge) were highly significant ( p < 0.001) for all five studied traits. The AMMI biplot selected G2 and G5 as the most stable genotypes, while the GGE biplot identified G1 and G5 as stable. Individual stability parameters: ASTAB (average stability table [or adaptation stability]), ASI (absolute stability index), AVAMGE (average absolute main genotype × environment interaction effect), DA, DZ, EV (averages of the squared eigenvector values), FA (factor analysis coefficient), MASI (modified absolute stability index), SIPC (sums of the absolute value of the IPCA scores), ZA (absolute value of the relative contribution of IPCAs to the interaction), WAAS (weighted average of absolute scores), Ecoval, and Shukla identified G2 and G5 as the most stable genotypes. HMRPGV (harmonic mean of the relative performance of the genotypic value), RPGV (Relative performance of the genotypic value), HMGV (harmonic mean of the genotypic value), GAI (geometric adaptability index), VAR (vector autoregressive), CV (coefficient of variation), and superiority index (Pi) ranked G5 and G1 as stable. The average rank sum of stability groups, AMMI statistics, and the WAAS index also ranked G2 and G5 as the top two. BLUP evaluations placed G5 first and G1 second, whereas parametric and nonparametric methods ranked G5 first and G2 second. Overall, G5 and G2 were selected as stable. The MTSI also selected G5 and G2, and hence, these genotypes are recommended for further verification and demonstration before being recommended for wider forage production in Tigray.