Abstract
Cowpea is an important multipurpose legume crop that used for food, feed and vegetable worldwide. Developing the high yield cultivars is the first target in cowpea breeding, however, the genetic basis of this complex trait is not yet well understood. To discover the genetic architecture of cowpea yield, a total of 215 cowpea landraces collected from Zhejiang Province were evaluated for four yield-related traits including branch number per plant (BNP), grain number per pod (GNP), pod length (PL), and pod number per plant (PNP). By resequencing this diversity panel, total of 3,880,169 high-confidence single nucleotide polymorphisms (SNPs) were identified, population structure analysis showed that these cowpea landraces were classified into four subpopulations and the subpopulation division was highly related to the pod length and pod-type. Through conducting a GWAS on the four traits, 24 genomic regions significantly associated with cowpea yield were detected and haplotype analysis showed the favorable genotypes of each locus has stronger genetic effect on the yield-related traits. Based on the cowpea G98 reference genome, six predicated genes (VuG9806G022730, VuG9809G015960, VuG9801G022820, VuG9801G008990, VuG9801G016500, VuG9807G013020) were identified as the likely candidate genes for BNP_6.2, BNP_9.1, GNP_1.1, PL_1.1, PNP_1.2 and PNP_7.1, respectively, which involving in multiple pathways such as auxin response and regulation, cell expansion and ovary development. These results will facilitate the molecular breeding of high yield cultivars in cowpea and benefit for improving the global food security and the nutritional structure of human diets. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s11032-025-01585-x.