High rate of calculation errors in mismatch distribution analysis results in numerous false inferences of biological importance.

摘要

One of the greatest challenges in unravelling the demographic history of populations is the lack of a universal molecular clock as first envisioned by Zuckerkandl and Pauling (1965) and our lack of certainty concerning even calibrated or local (within and between closely related species) molecular clocks (Hickerson et ah, 2003). Evolutionary rates can vary widely within and among groups of organisms (for example, Garcia-More-no, 2004; Thomas et ah, 2006). The hypothesised causes of this variation are diverse and include varying population sizes and selection pressures, different generation times and the influence of body size on metabolic rates (for example, Baer et ah, 2007). The stochasticity of the coalescence process within lineages (Edwards and Beerli, 2000) adds even more complexity to simple uniform calibrations. Although attempts to overcome or account for these uncertainties remain at the forefront of evolutionary research, our focus in this study is on a much more mundane problem. Very simply, assuming a given substitution or divergence rate, are authors correctly applying it in standard algorithms in order to infer the timing of demographic events?