Potential problems in estimating the male-to-female mutation rate ratio from DNA sequence data

It is commonly believed that the rate of mutation is much higher in males than in females because the number of germ-cell divisions per generation is much larger in males than in females. However, the precise magnitude of the male-to-female mutation rate ratio (α _m ) remains unknown. Recently there have been efforts to estimate α _m by using DNA sequence data from different species. We have studied the potential problems in such an approach. We found that the rate of synonymous substitution varies about fivefold among X-linked genes, as large as the variation among autosomal genes. This large variation makes the assumption of selective neutrality of synonymous changes dubious, so one should be cautious in using the synonymous rates in X-linked and autosomal genes to estimate α _m . A similar difficulty was also observed in using nonhomologous intron sequences to estimate α _m . Contrary to the expectation that X-linked sequences should evolve more slowly than autosomal sequences, theAlu repeat in the last intron of the X-linked zinc finger gene has evolved faster than the four autosomalAlu repeats used in this study. It appears that the best way to estimate α _m is to use homologous sequences. However, such sequences may be involved in gene conversion events. In fact, we found evidence that the Y-linked and X-linked zinc finger genes have been involved in multiple conversion events during primate evolution. Thus, the possibility of gene conversion should be considered when using homologous sequences to estimate α _m . Presented at the NATO Advanced Research Workshop onGenome Organization and Evolution, Spetsai, Greece, 16–22 September 1992