Population structure and diversity of cowpea ( Vigna unguiculata [L.] Walp.): Genomic inference and preliminary distribution of breeding-relevant traits in a Niger-centred Sahelian diversity panel

The study of genetic diversity is essential for understanding population structure and optimizing breeding programs. To support cowpea breeding efforts, we conducted a diversity study on a panel of 185 accessions, including 125 newly collected accessions from Niger and additional accessions provided by research institutions in Nigeria, Burkina Faso, Senegal and Niger. This study aimed to analyse cowpea population structure and genomic diversity and to provide a preliminary assessment of phenotypic variation across genomic clusters. Genotyping-by-sequencing generated 20,918 single nucleotide polymorphisms (SNPs), which were used to assess population structure and genetic diversity. Population structure was inferred using a Bayesian approach (ADMIXTURE), and the resulting clustering pattern was corroborated by a multivariate method (DAPC). Seven genetic clusters were identified (A1–A7). Accessions from Niger were predominantly assigned to clusters A1, A2, A3 and A5; those from Senegal were mainly assigned to A4, whereas accessions from Nigeria were predominantly assigned to A7. Strong genetic differentiation was observed between accessions from Nigeria and those from Niger. The evaluation of phenological traits and yield components revealed marked differences among clusters. Most clusters showed low and multimodal grain and husk yield distributions, whereas cluster A5 consistently combined early maturity, high haulm yield and superior reproductive performance, highlighting its breeding potential. Overall, this study generated a large SNP marker dataset and provided integrated genomic and preliminary phenotypic information for a diverse cowpea germplasm panel. These resources will be useful for further studies of population genetics, breeding and association mapping in cowpea, an important legume crop for Sahelian countries.