Skip to main content
Fig. 1 | BMC Evolutionary Biology

Fig. 1

From: Expression and regulatory asymmetry of retained Arabidopsis thaliana transcription factor genes derived from whole genome duplication

Fig. 1

Linear model of the degree of duplicate retention in function groups based on genes features. a Relationships between gene counts and odds of retention of WGD duplicates across functional groups (α = orange, β = green, γ = blue). The correspondence between group sizes (numbers of genes) and degrees of retention (odds ratios) was determined using the square of the Pearson product-moment correlation coefficient (R2, α = 0.05, β = 0.16, γ = 0.04). b A heatmap of the Pearson product-moment correlation coefficient (PCC) between the values of a feature across different function groups (rows) and the odds of retention of functions groups from a particular WGD event (columns, indicated by the symbols α, β, and γ). Darker red: stronger positive correlation. Darker blue: stronger negative correlation. Features with different sign of correlation across WGD events are indicated by black arrows. Features with a large (≥0.20) difference in PCCs with the same sign are indicated by open arrows. c The observed odds of duplicate retention (x-axis) for each group plotted against the predicted odds of retention (y-axis) from the best model for each event (α = orange, β = green, γ = blue). Dotted line: equality between predicted and observed retention odds. Values from TFs are indicated by a black arrow while values from protein kinases are indicated by an open arrow. Red dot (TFγ’): the predicted odd ratio for TFs from the γ event after adjusting for difference in percent identity of TF genes. Performance of the models was assessed by calculating the R2 between the observed and predicted odds ratio for each event (α = 0.87, β = 0.83, γ = 0.65)

Back to article page