Genome-wide association study of yield-component traits in US upland cotton (Gossypium hirsutum L.)

Abstract

Genetic improvement of cotton (Gossypium spp.) yield is challenged by the complex and polygenic nature of yield and its underlying component traits. In this study, an elite diversity panel of 348 upland cotton accessions, genotyped using the CottonSNP63K array, was evaluated across 10 environments in the US Cotton Belt for four yield-component traits: lint percent (LP: %), boll weight (BW: g), seed index (SI: g), and lint index (LI). All traits exhibited high broad-sense heritability (>60%), and all except for SI showed high genomic heritability. Positive correlations were observed among BW, SI, and LI, whereas SI was negatively correlated with LP. Genome-wide association study of the four traits identified 308 significant marker–trait associations (MTAs), including 80 stable MTAs distributed across 19 genomic regions, of which 15 are novel associations. These 19 genomic regions included MTAs detected in at least three environments and explained 7.23%–11.46% phenotypic variation. Notably, this study confirmed the pleiotropic region influencing SI and BW on chromosome D11 and identified a novel co-localized group on D12 associated with LP and LI. Haplotype analysis further revealed superior allele combinations associated with LP, showing a high trait value (LP: 39.77%). The genomic regions and favorable haplotypes identified in this study provide valuable resources for harnessing genetic diversity, accelerating targeted genetic improvement, and facilitating the gene discovery of yield traits in cotton.