Multi-model genome-wide association analysis of agronomic traits in a Kabuli chickpea MAGIC-subset population (Cicer arietinum L.) across Mediterranean environments
Chickpea is one of the most consumed legumes due to its high nutritional value and accessibility to low-income populations. However, due to climate change, chickpea cultivation is exposed to various environmental stresses affecting its production and productivity. This study evaluates the agronomic performance of 168 MAGIC subset population across two Mediterranean environments, in Marchouch (Morocco) and Terbol (Lebanon). Genome-wide association studies (GWAS) were conducted to identify potential marker-trait associations (MTAs) using the general linear model (GLM), the mixed linear model (MLM), and the fixed- and random-effect circulant probability unification (FarmCPU), with kinship and principal components used as covariates. The results revealed high genetic variation among the genotypes tested, with significant genotype-by-environment interactions for most traits studied. Genotypes with good agronomic performance (M-1407, M-2038, M-2079, M-242, M-2551, and M-987) were identified under both environments. Early flowering and maturation resulted in a significant increase in grain yield of around 66%. Grain yield varied from 151.85 to 882.7 g m-2 and from 183.59 to 365.76 g m-2 under Marchouch and Terbol conditions, respectively, showing higher genetic variation under Marchouch than under Terbol. Correlation analysis revealed strong, significant correlations among the studied traits. GWAS revealed clear genetic variation across environments, with Marchouch showing stronger and more consistent association signals than Terbol. In total, 72 reliable MTAs were detected for phenological traits, 38 for plant height, 18 for grain yield, and 197 for hundred-seed weight across both sites. A major genomic hotspot on chromosome 4 (11.97–13.63 Mbp) harbored stable and pleiotropic MTAs. Functional annotation of regions surrounding significant SNPs revealed 59 putative candidate genes, highlighting potential biological processes related to growth, signaling, and stress responses that require further validation. These results provide a foundation for further research on marker-assisted selection to improve chickpea productivity and yield stability in stressed environments.