The boll opening rate (BOR) is an early maturity trait that plays a crucial role in cotton production in China, as BOR has a significant effect on defoliant spraying and picking time of unginned cotton, ultimately determining yield and fiber quality. Therefore, elucidating the genetic basis of BOR and identifying stably associated loci, elite alleles, and potential candidate genes can effectively accelerate the molecular breeding process. In this study, we utilized the mixed linear model (MLM) algorithm to perform a genome-wide association study (GWAS) based on 4,452,629 single-nucleotide polymorphisms (SNPs) obtained through whole-genome resequencing of a natural population of 418 upland cotton accessions and phenotypic BOR data acquired from five environments. A total of 18 SNP loci were identified on chromosome D11 that are stable and significantly associated with BOR in multiple environments. Moreover, a significant SNP peak (23.703–23.826 Mb) was identified, and a GH-D11G2034 gene and favorable allelic variation (GG) related to BOR were found in this genomic region, significantly increasing cotton BOR. Evolutionary studies have shown that GH-D11G2034 may have been subjected to artificial selection throughout the variety selection process. This study provides valuable insights and suggests that the GH-D11G2034 gene and its favorable allelic variation (GG) could be potential targets for molecular breeding to improve BOR in upland cotton. However, further research is needed to validate the function of this gene and explore its potential applications in cotton breeding programs. Overall, this study contributes to the advancement of genetic improvement in early maturity and has important implications for the sustainable development of the cotton industry.

Upland cotton (Gossypium hirsutum L.) dominates Xinjiang’s cotton production, but the narrow genetic base of modern cultivars limits further yield gains. Here, 356 Xinjiang upland cotton accessions (184 northern, 172 southern) were genotyped at 280 elite trait locus–trait combinations (270 unique markers) using a liquid-phase chip. Pairwise kinship was inferred with the KING-robust estimator and cross-validated by IBS similarity, population structure was examined by PCA, and a marker-weighted comprehensive elite genotype index (S) was constructed. The overall missing rate was 0.43%, and five identical and 101 near-identical accession pairs were detected. PC1 and PC2 explained only 9.81% and 7.29% of the variation, with substantial regional overlap. Index S ranged from 0.39 to 0.57, led by Tkuo, Y80, Ba 44435, Ziyu R040068 and Youzhi 5; salt tolerance (0.07), lint yield (0.14), seed index (0.19) and fiber strength (0.22) showed the lowest carriage of the marker-defined elite alleles scored here. After excluding close relatives, 62,252 crosses (98.52%) satisfied the constraint, and 30 crosses (19 inter-regional) were recommended, the best being Ziyu TY112 × Tkuo (joint coverage U = 0.65). These parents and crosses provide candidates for parental selection, pending field validation.

Upland cotton (Gossypium hirsutum L.) dominates Xinjiang's cotton production, but the narrow genetic base of modern cultivars limits further yield gains. Here, 356 Xinjiang upland cotton accessions (184 northern, 172 southern) were genotyped at 280 elite trait locus–trait combinations (270 unique markers) using a liquid-phase chip. Pairwise kinship was inferred with the KING-robust estimator and cross-validated by IBS similarity, population structure was examined by PCA, and a marker-weighted comprehensive elite genotype index (S) was constructed. The overall missing rate was 0.43%, and five identical and 101 near-identical accession pairs were detected. PC1 and PC2 explained only 9.81% and 7.29% of the variation, with substantial regional overlap. Index S ranged from 0.39 to 0.57, led by Tkuo, Y80, Ba 44435, Ziyu R040068 and Youzhi 5; salt tolerance (0.07), lint yield (0.14), seed index (0.19) and fiber strength (0.22) were the main deficiencies. After excluding close relatives, 62,252 crosses (98.52%) satisfied the constraint, and 30 crosses (19 inter-regional) were recommended, the best being Ziyu TY112 × Tkuo (joint coverage U = 0.65). These parents and crosses provide candidates for parental selection, pending field validation.