dN(eutralist) > dS(electionist)? Part 2

Last week’s post dealt with the debate over differences in the efficacy of purifying selection across human genomes. This week, we’ll look at the differences in de novo mutation rates across populations. The human de novo mutation rate has gone through a lot of recalibrations over the last few years (2.5 x 10-8 mutations per nucleotide per generation – Nachman and Crowell (2000) to 1.2 x 10-8 mutations per nucleotide per generation – Scally and Durbin (2012), Campbell et al. (2012), etc.), with plenty of variation with types of loci, male versus female germlines, and mutation types.

The de novo mutation rate is influenced by environmental changes resulting in DNA damage. The extent to which these de novo mutations arise, and are maintained in a population are also intertwined with (a) the population’s demography, and (b) the selection regimes at these mutated, and linked loci. A comprehensive study of de novo mutations (compared to the ancestral population) that are fixed (also called ‘private alleles’) in a population, but are still segregating in a third population should thus reveal the local dynamics of novel mutations.


Overrepresentation of TCC->TTC within Europe, Figure 1 from Harris (2015) – http://dx.doi.org/10.1073/pnas.1418652112

Harris (2015) in a recent study of the variation in the frequencies of de novo mutations across the Phase 1 of the 1000 Genome Project data reports elevated mutation rates of a certain class (TCC->TTC) of non-synonymous substitutions in Europeans, when compared to Asians and Africans, which have often been linked to UV related damage. Noticeably, this also coincides with the evolution of lighter skin in Europeans after diversification outside of Africa, although no direct correlates (or positive selection for this class of mutations) between the two phenomena have been established or studied. In general, her results indicate (1) elevated C->T transitions in all Europeans, particularly in a class of TCC->TTC, (2) this ‘change’ in mutation rate occurred in the common ancestor of all Europeans, after divergence out of Africa, but prior to diversification inside Europe, around 25,000-60,000 ybp, (3) the non-bioinformatic (i.e. not due to erroneous detection methods) cause of this observation, and (4) a bias in transcription-coupled repair mechanism in “correcting” this class of mutations in European populations.

Some interesting questions arise as a consequence of these results, which feed well into our continuing debate about the dynamics of novel mutations.

  • Seeing that the (1) elevated levels of TCC->TTC mutations in Europeans occurred as a result of some environmental change (here excessive UV damage), (2) there is a bias in the transcription-coupled repair mechanism in correcting these mutations in Europeans, and (3) that these mutations are fixed in the population, one would surmise that this is a definitive consequence of the demography on European populations?
  • The observation that there isn’t any difference in the efficacy of purifying selection across Europeans, and Asians (as reported by Do et al. (2015)), is conditional on the same mutation rate across Europeans and Asians, which as Harris reports, isn’t true across all classes of loci. So perhaps demography (say a population bottleneck, or expansion) did play a role in the efficacy of purifying selection after all?

Feel free to add your comments!


Harris, K. Evidence for recent, population-specific evolution of the human mutation rate, Proceedings of the National Academy of Sciences, DOI:http://dx.doi.org/10.1073/pnas.1418652112


About Arun Sethuraman

I am a computational biologist, and I build statistical models and tools for population genetics. I am particularly interested in studying the dynamics of structured populations, genetic admixture, and ancestral demography.
This entry was posted in evolution, mutation, theory and tagged , , , . Bookmark the permalink.