Table 1.
Observed and expected levels of unbiased heterozygosity in sunflower groups.
Figure 1.
STRUCTURE plot of the sunflower association mapping panel with K = 3 clusters based on all polymorphic SNP markers.
The plot is sorted according to line classification. I-HA = INRA-HA, HA-NO = HA-non-oil, HA-O = HA-oil, I-RHA = INRA-RHA, RHA-NO = RHA-non-oil, RHA-O = RHA-oil, The four “Other” categories refer to non-oil lines, oil lines, open pollinated varieties and landraces, and introgressed lines, respectively.
Figure 2.
Heat map of linkage disequilibrium across the sunflower genome.
Individual data points reflect squared allele frequency correlations (r2) for all possible pairs of polymorphic SNP markers, MAF ≥10%. The x- and y-axes correspond to the 17 linkage groups in sunflower with marker orders based on the work of Bowers et al. [21]. Note that the values above and below the diagonal are identical.
Figure 3.
Sliding window analysis of squared allele frequency correlations (r2) across the sunflower genome.
Figure 4.
Population genetic differentiation (FST) between heterotic groups (RHA vs. HA lines) for polymorphic SNPs, MAF ≥10%, plotted against genetic map position. The red colored dots represent individual SNPs that showed evidence of divergence due to selection in the outlier analysis.
Figure 5.
Manhattan and quantile–quantile plots of branching associations.
(A, B, C) Upper panels: Manhattan plots of branching associations in three locations (GA, IA, BC respectively) plotted for the three models tested: red = K, blue = P+K, dark grey = Q3+K. The dashed line indicates the significance threshold based on the multiple testing correction method of Gao et al. [35] (alpha = 0.05, P = 0.00025, log 1/P = 3.60). Lower panels: Quantile-quantile plots of branching associations in all three locations plotted for the three models tested.
Figure 6.
Manhattan and quantile–quantile plots of flowering time associations.
(A, B, C) Upper panels: Manhattan plots of flowering (DTF) associations in three locations (GA, IA, BC respectively) plotted for the three models tested: red = K, blue = P+K, dark grey = Q3+K. The dashed line indicates the significance threshold based on the multiple testing correction method of Gao et al. [35] (alpha = 0.05, P = 0.00025, log 1/P = 3.60). Lower panels: Quantile-quantile plots of flowering time associations in all three locations plotted for the three models tested.
Table 2.
Summary of significant branching associations.
Table 3.
Summary of significant flowering time (DTF) associations.
Figure 7.
Graphical genotypes from linkage group 10.
Graphical genotypes from linkage group 10 plotted against a heat map of the branching data sorted by average level of branching across all three locations. Unbranched plants (red) are to the left whereas highly branched plants (green) are to the right. Note that the scale of the y-axis scale changes based on marker density, and white squares in the phenotype heat map represent missing data in an individual location.