Local similarity in evolutionary rates extends over whole chromosomes in human-rodent and mouse-rat comparisons: Implications for understanding the mechanistic basis of the male mutation bias

The sex chromosomes and autosomes spend different times in the germ line of the two sexes. If cell division is mutagenic and if the sexes differ in nu mber of cell divisions, then we expect that sequences on the X and Y chromo somes and autosomes should mutate at different rates. Tests of this hypothe sis for several mammalian species have led to conflicting results. At the s ame time, recent evidence suggests that the chromosomal location of genes o n autosomes affects their rate of evolution at synonymous sites. This sugge sts a mutagenic source different from germ cell replication. To correctly i nterpret the previous estimates of male mutation bias, it is crucial to und erstand the degree and range of this local similarity. With a carefully cho sen randomization protocol, local similarity in synonymous rates of evoluti on can be detected in human-rodent and mouse-rat comparisons. However, the synonymous-site similarity in the mouse-rat comparison remains weak. Simula tions suggest that this difference between the mouse-human and the mouse-ra t comparisons is not artifactual and that there is therefore a difference b etween humans and rodents in the local patterns of mutation or selection on synonymous sites (conversely, we show that the previously reported absence of a local similarity in nonsynonymous rates of evolution in the human-rod ent comparison was a methodological artifact). We show that Link-age effect s have a long-range component: not one in a million random genomes shows su ch levels of autosomal heterogeneity. The heterogeneity is so great that mo re autosomes than expected by chance have rates of synonymous evolution com parable with that of the X chromosome. As autosomal heterogeneity cannot be owing to different times spent in the germ line, this demonstrates that th e dominant determiner of synonymous rates of evolution is not, as has been conjectured, the time spent in the male germ line.