Figure 1

Plant genomes maintain homogeneous distributions of protein domain architectures. The categories of protein domain architectures, labeled on top of each histogram and on bottom of each probability plot, are overall predicted (the left panel), species-unique (second to the left panel), as well as single-domain, double-domain, triple-domain and equal to or greater than four-domain architectures (the right four panels). The numbers are mean ± standard error. The x-axis for both the upper and lower panels is the proportions of Pfam-predicted domain architectures per genome in each category. The upper panel shows frequency distributions of the percentages of these categories of domain architectures. The lower panel is the probability plot (5% significance level) of the percentages of these various categories of domain architectures across plant species. The y-axis is the probability distributions relative to the mean values. AD represents the value of Anderson-Darling normality test. Note that the proportions of domain architectures of equal to or greater than four domains do not follow a normal distribution as evidenced by the associated p-value in the probability plot.