population expansion

Keinan A & Clark AG 2012 Recent explosive human population growth has resulted in an excess of rare genetic variants. Science 336:740-743.

  • for European history, these studies estimated as much as 0.5% growth in effective population size per generation since the split of the ancestors of Europeans and East Asians ~1000 generations ago, resulting in an effective population size of a few tens of thousands today
  • these studies did not capture the full scope of human expansion, which may be due to the models not allowing for a recent acceleration in growth rate (5)
  • the limited sample size of these studies (at most 60 individuals), which only allowed capturing variants of frequency as low as ~1% in the sample, has provided a limited view of rare variants in the population
  • rare variation adds information on mutations that have occurred during recent epochs of accelerated explosive growth and that may be identified from sequencing larger sample sizes
  • several ongoing projects are sequencing ever-larger numbers of individuals genome-wide
  • learning from the frequencies of genetic variants about the demographic history during the past 10,000 years requires capturing variants that entered the population in that time span
  • many such variants are likely to be rare in the population as a whole (i.e., frequency <0.1%; see below), requiring sequencing of a larger sample of individuals than previously considered in demographic studies
  • it is also necessary to address how assumptions underlying most population genetic analysis tools are violated with samples of thousands or more individuals
  • this may require generalizations of population genetics theory to cases in which the sample size is not necessarily smaller than the historical effective population size
  • considering a sample of 10,000 individuals, the number of singletons estimated was at least 5 times that predicted by the standard Wright-Fisher model (18)
  • the expected number of doubletons was at least 3 times the prediction
  • the model estimates pointed to a recent extreme growth of 9.4% per generation [95% Bayesian credible interval (CI), 4.5 to 14.5%], starting 1400 years ago (95% CI, 900 to 2800 years ago)
  • the current global human population growth rate is estimated at 1.1% per year, equivalent to about 30% per generation
  • the present effective population size of Europeans as estimated by the model is 1.1 million (95% CI, 0.3 to 1.9 million) (18)
  • a larger sample size allows identification of rarer polymorphisms that are, on average, due to more recent mutations
  • the larger the sample size, the more recent the epoch it probes
  • the excess of rare variants found with a large sample size predicts a growth of 5 to 14% per generation over the past 900 to 2800 years, compared to models of smaller sample size that predicted growth of between 0.2 and 0.7% over a longer period of 20,000 to 30,000 years (5, 6)
  • we contrasted three simplified models of population history to exemplify the general effects of recent explosive growth and sample size on the SFS
  • the expected SFS under the three models (Fig. 2) points to considerable inflation in the fraction of variants that are singletons when recent growth is introduced in model 3
  • Fig. 2
  • (i) a population that has been of constant size throughout history
  • (ii) a model previously fit to the derived allele frequency spectrum of Europeans, which includes an out-of-Africa population bottleneck and a second, more recent, population bottleneck (21)
  • (iii) the same two-bottleneck model of European history with the addition of recent exponential growth from a population size of 10,000 at the advent of agriculture to an extant effective population size of 10,000,000, which amounts to 1.7% growth per generation during the last 400 generations
  • explosive growth also results in a haplotype structure that deviates from the prediction of standard population genetics
  • recombination will not have broken down linkage disequilibrium between rare mutations and neighboring common variants