Fig. 1.

—Duplicated genes persist in genome evolutionary hotspots. (a) We estimated the mean nonsynonymous nucleotide substitutions per nonsynonymous site (d_N: black column) and the synonymous substitutions per synonymous site (d_S; white column), for the three singleton genes (Locus tag genes in the white rectangles) immediately flanking a duplicated gene (black rectangle) at either side. (b) The mean d_N for each region containing a duplicate (GRDs) within each chromosome was calculated and compared to that of genome regions containing only singletons (GRSs). We identified seven chromosomes in which GRDs exhibited significantly higher d_N than GRSs (red-labeled roman numbers in x axis). (c) For each of the duplicates contained in each GRD of each chromosome, we searched for its paralogue elsewhere in the genome. Then we compared the d_N of both these groups and found that when one GRD of a chromosome exhibited a mean d_N below the mean d_N for GRSs (white boxes), their paralogs exhibited the inverse pattern (gray boxes), and vice versa. Red-labeled chromosomal numbers in the x axis indicate those for which evidence exist that at least one of the paralogs is in a GRD with high d_N.