1. Howard Hughes Medical Institute, Whitehead Institute, and Department of Biology, Massachusetts Institute of Technology, 9 Cambridge Center, Cambridge, Massachusetts 02142, USA

Correspondence to: David C. Page Correspondence and requests for materials should be addressed to D.C.P. (e-mail: Email: dcpage@wi.mit.edu).

Abstract

In 1947, it was suggested that, in humans, the mutation rate is dramatically higher in the male germ line than in the female germ line¹. This hypothesis has been supported by the observation that, among primates, Y-linked genes evolved more rapidly than homologous X-linked genes^{2, 3, 4, 5, 6}. Based on these evolutionary studies, the ratio (_m) of male to female mutation rates in primates was estimated to be about 5. However, selection could have skewed sequence evolution in introns and exons^{7, 8, 9, 10}. In addition, some of the X–Y gene pairs studied lie within chromosomal regions with substantially divergent nucleotide sequences^{7, 11, 12}. Here we directly compare human X and Y sequences within a large region with no known genes. Here the two chromosomes are 99% identical, and X–Y divergence began only three or four million years ago, during hominid evolution^{13, 14, 15}. In apes, homologous sequences exist only on the X chromosome. We sequenced and compared 38.6 kb of this region from human X, human Y, chimpanzee X and gorilla X chromosomes. We calculated _m to be 1.7 (95% confidence interval 1.15–2.87), significantly lower than previous estimates in primates. We infer that, in humans and their immediate ancestors, male and female mutation rates were far more similar than previously supposed.