Multi-environments and multi-models association mapping identified candidate genes of lint percentage and seed index in Gossypium hirsutum L.

Upland cotton ( Gossypium hirsutum L.) accounts most of the natural fiber production worldwide. Lint percentage (LP) and seed index (SI) are important components of cotton fiber yield, which is a constant breeding goal of cotton. So, the loci underpinning LP and SI should be extensively dissected. Here, one single-locus and four multi-locus genome-wide association study (GWAS) models were employed to detect candidate loci for lint percentage and seed index under seven environments with 196 upland cotton accessions and 41,815 single nucleotide polymorphism (SNP) markers. Totally, 39 and 45 significant quantitative trait locus (QTL) were identified in at least two environments or two models, including 24 previously reported QTLs and six pleiotropic QTLs. Referred to the genome and gene expression database of TM-1, 614 candidate genes were detected for lint percentage and seed index, including 103 genes preferentially expressed in fiber or ovule. The gene Gh_A10G0378, functioned in potassium ion transport, was considered to be related to lint percentage. Collectively, the associated markers and promising genes detected herein will help to elucidate the genetic architecture of lint percentage and facilitate fiber yield improvement in cotton.