The rate of adaptive evolution in animal mitochondria

Abstract

We have investigated whether there is adaptive evolution in mitochondrial DNA, using an extensive data set containing over 500 animal species from a wide range of taxonomic groups. We apply a variety of McDonald-Kreitman style methods to the data. We find that the evolution of mitochondrial DNA is dominated by slightly deleterious mutations, a finding which is supported by a number of previous studies. However, when we control for the presence of deleterious mutations using a new method, we find that mitochondria undergo a significant amount of adaptive evolution, with an estimated 26% (95% confidence intervals: 5.7-45%) of nonsynonymous substitutions fixed by adaptive evolution. We further find some weak evidence that the rate of adaptive evolution is correlated to synonymous diversity. We interpret this as evidence that at least some adaptive evolution is limited by the supply of mutations.

Keywords: adaptive evolution; distribution of fitness effects; genetic diversity; mitochondria; mutation rate; site frequency spectrum.

Figure 1

Histogram showing the frequency distribution of DoS values for the data set.

Figure 2

Graph of our estimate of α (using the Messer–Petrov method) plotted against the frequency category of polymorphism. An asymptotic curve of the form y = a + b e ^−cx was fitted to the data in order to obtain an estimate of α.

Figure 3

Bar chart of observed and expected values of p _n/p _s, as predicted from fitting a distribution to the DFE to the complete data set, plotted against the frequency category of polymorphism.