One thing men are really good at is making mistakes—just ask any woman. When it comes to mutations we are ten times better than women at ensuring the evolution of the species.
Knowing the actual rate of mutation in humans—or any other species—is important for many reasons. For one thing, it tells us about the maximum possible rate of evolution. For another, it gives us an important clue about the differences between beneficial, detrimental, and neutral alleles.
It's a lot more difficult to measure mutation rates than you might imagine. In theory, you could sequence the genomes of hundreds of parents and their offspring and identify mutations that must have occurred in the germ lines of the parents. In practice, this is far too expensive and time-consuming and, besides, it will miss any severely detrimental mutations.
But let's say you did the experiment in spite of the time and money. If the measured mutation rate turned out to be close to the calculated value, then you could assume that most of the mutations were neutral. A few might be beneficial.
Another possibility is to measure the number of differences between two individuals who are separated by a large number of generations. In this case you are measuring the combined effect of mutation and the fixation of alleles in a population. This is what we do whenever we compare gene sequences from different species.
Alleles can be fixed by natural selection or by random genetic drift. If most are fixed by natural selection (adaptation) then you'll learn very little about the overall mutation rate aside from a minimum estimate. That's because you don't know the fitness of every allele and how fast it became fixed in the population and you don't know how many mutations were detrimental or neutral, and what happened to them.
Calculating the rate of evolution in terms of nucleotide substitutions seems to give a value so high that many of the mutations must be neutral ones.
Motoo Kimura (1968)
However, you have a fighting chance if most mutations give rise to neutral alleles. In that case, the overall rate of fixation by random genetic drift is the same as the mutation rate [see: Random Genetic Drift and Population Size]. The data suggest that this is the correct scenario. When we compare individual genes from different species, the observed differences are consistent with the expected result if most of the differences are due to the fixation of neutral alleles by random genetic drift.
For the comparison between humans and chimpanzees, the estimated rates are remarkably consistent. They range from about 2 × 10-8 to about 5 × 10-8 mutations per nucleotide (base pair) per generation (Nachman, 2004; Britten, 2002). This agrees with the known error rate of DNA replication, which is about 10-10 per nucleotide per replication. Since there are about 400 DNA replications between the male zygote and mature sperm, this translates to 4 × 10-8 mutations per nucleotide per generation [see, Mutation Rates].
This is where men come in. There are many fewer cell divisions in the female line—about 30—so the egg contributes fewer mutations than the sperm. In fact, for most purposes we can ignore women in these calculations. Men have another big advantage. They have a Y chromosomes that's passed down directly from father to son and it doesn't recombine with any female chromosomes.1 You don't need to worry about fixation.
If you sequence Y chromosomes from related men you can get a direct estimate of the mutation rate provided most of the alleles are neutral. It's best to choose men who are distantly related since there won't be many differences between closely related men. Two sons, for example, are likely to have identical Y chromosomes.
Xue et al. (2009) did the experiment [Human mutation rate revealed]. They sequenced the Y chromosomes of two men who were separated by 13 generations. After eliminating repetitive regions, the relevant region of comparison was 10.15 × 106 nucleotides (base pairs, 10.15 Mb). The men differ at four confirmed sites. This gives a mutation rate of 3.0 × 10-8 per generation or 0.75 × 10-10 per nucleotide per DNA replication.
The agreement is remarkable. What this means is that we have a good handle on the mutation rate in humans and we have growing evidence that most mutations are neutral (i.e. most of our genome is junk).
1. This isn't strictly correct but you can ignore the small regions where recombination is possible.
Britten, R.J. (2002) Divergence between samples of chimpanzee and human DNA sequences is 5%, counting indels. Proc. Nat. Acad. Sci. USA 99:13633-13635. [doi: 10.1073/pnas.172510699]
Nachman, M.W. (2004) Haldane and the first estimates of the human mutation rate. J Genet. 83:231-233. [PubMed]
Xue, Y., Wang, Q., Long, Q., Ng, B.L., Swerdlow, H., Burton, J., Skuce, C., Taylor, R., Abdellah, Z., Zhao, Y.; Asan, Macarthur, D.G., Quail, M.A., Carter, N.P., Yang, H., Tyler-Smith, C. (2009) Human Y Chromosome Base-Substitution Mutation Rate Measured by Direct Sequencing in a Deep-Rooting Pedigree. Curr Biol. Aug 26. [Epub ahead of print] [doi: 10.1016/j.cub.2009.07.032]
Xue, Y., Wang, Q., Long, Q., Ng, B., Swerdlow, H., Burton, J., Skuce, C., Taylor, R., Abdellah, Z., & Zhao, Y. (2009). Human Y Chromosome Base-Substitution Mutation Rate Measured by Direct Sequencing in a Deep-Rooting Pedigree Current Biology DOI: 10.1016/j.cub.2009.07.032