Fig. 4: FH-based GWAS identified key QTGs associated with agronomic traits and revealed the genetic basis of variety improvement in cotton.

a Comparison of the numbers of QTGs and QTLs associated with the phenotype. b Proportion of overlapped QTLs and QTGs according to flanking region size. c Distribution of QTGs on the sub-genome. d Relationship of the relative number of superior FHs (left panel), average effects of all significant FHs (right panel), and phenotypes. Each point represents a sample, point size represents the average effect size of significant FHs. e Relative number of superior FHs possessed by cottons registered or introduced in different time periods. n is the sample size. f Genetic basis of phenotypic correlations in cotton. Pie charts plot the proportion of FHs with significant pleiotropy effects on FL, FS, and FS, LP (upper panel), and boxplots show the absolute effect sizes for FHs with inverse or isotropic pleiotropic effects (lower panel), n is the number of genes with corresponding pleiotropic effects, two-sided student t-test are used for statistical inference (**p < 0.01, ns not significant). In a boxplot, the center line represents the median, the lower and upper box hinges corresponding to the first and third quantiles, whiskers extend to the minimum and maximum values, and individual points beyond the whiskers (if present) denote outliers. Source data are provided as a Source Data file.