Feedback archiveFeedback 2016

Mutation accumulation rates are consistent with biblical creation


J.P. writes in from the UK:

Have you or anyone at CMI a response to this work by a PhD student who claims to have solved the mtDNA mutation rate problem for the evolutionists against us?

New 'Molecular Clock' Aids Dating Of Human Migration History.1

I would be very interested to have some expert wisdom if you or your Australian colleagues can advise. I notice it is 6 years old, so presumably someone has assessed the work.

Yours in the Lord’s work,

CMI’s Dr Robert Carter responds:


Yes, this is old material, but no, I don't see that anyone has addressed it directly so I will attempt to do so for your (and my) edification.

The basic idea is that a trait under positive selection will accumulate more mutations over time than it should (because selection ‘drives’ it forward). On the other hand, traits under purifying selection will display lower mutation rates because most mutations to this trait/gene are harmful and are therefore being removed over time. The latter is the main thrust of the paper under discussion. When one takes these things into account, the absolute age of a branch in a phylogenetic tree might be affected.

Figure 1: Modelling mutation accumulation over time vs. population size in Mendel’s Accountant. The results from six different models are presented here. Each ran for 10,000 generations with a mutation rate of 0.1. All other parameters were set to program defaults. Population sizes were 10, 50, 100, 1000, 5000, and 10000. At the end of each model run I simply took the average number of mutations that appeared in each individual. In smaller populations, the mutation accumulation rate is approximately what would be expected with no drift or selection. But even in larger populations, where selection is much more effective, the number of accumulated mutations was not more than 20% less.

However, there is a limit to the amount of difference one can get between the measured, real-time mutation rate and the phylogenetic accumulation of mutations over time. I used Mendel’s Accountant2 to perform hundreds of population models and the best I could get was about a 20% difference, and only in large populations that had enough “spare” people to select away without threat of population extinction. In figure 1, the mutation rate is such that about 1000 mutations would be expected per person, but you can see that only about 800 have accumulated in the individuals in the larger populations. There are no error bars on this plot because I pulled a single example for each population size from among the many model variants I ran. Based on the shape of the curve, however, I am not very concerned that additional sampling will appreciably change the results.

This new argument has little to do with the creation/evolution debate. There is no great difference between the real mutation rate and the number expected to accumulate over time. It is just a matter of fine tuning the rate to get a date, after assuming a specific selection coefficient for the average mutation. I allowed for stronger selection than one would normally expect in order to get the results above, so the real-world difference between the observed and expected number of mutations accumulated should be less than this. But note that the population events discussed in the article are strongly dependent on carbon-14 dates, which most creationists would say are skewed downward the closer one gets to the Flood. Thus, there is a great deal of ‘art’ applied to the ‘science’ of generating these dates.

Robert Carter

Published: 3 December 2016

References and notes

  1. University of Leeds. “New ‘Molecular Clock’ Aids Dating Of Human Migration History.” ScienceDaily, 22 June 2009; sciencedaily.com/releases/2009/06/090604124023.htm. Return to text.
  2. Sanford, J., Baumgardner, J., Brewer, W., Gibson, P., and ReMine, W., Mendel’s Accountant: A biologically realistic forward-time population genetics program, SCPE 8(2):147–165, 2007. Springer-Verlag, Berlin, Heidelberg, pp. 386–392. (PDF, the program is available for download here.) Return to text.

Helpful Resources