Vincent Torley tries to understand fixation

I'm having an interesting discussion with some creationists. They claimed that the differences between the sequences of the human and chimpanzee genomes could not be explained by evolution. Therefore, it had to be due to design.

I wrote up a little post showing that there were about 44 million differences and that they could be accounted for by our understanding of population genetics and Neutral Theory. What his means is that the creationist explanation has to account for the fact that the vast majority of differences look like what we would expect if most of them were neutral and population genetics is correct. It's not good enough to simply invoke design and magic to explain the differences, you have to account for all the data.

Vincent Torley attempted to understand modern evolution theory. Some of these concepts were quite new to him because he doesn't have much of a biology background. We've had an exchange of posts were he expressed his astonishment and I try to explain evolution. You can find the links at: Vincent Torley apologizes and claims that he is not a liar.

I was thinking that this exchange would wind down but I was wrong. Vincent Torley is having second thoughts about accepting my explanation of Neutral Theory, population genetics, and mutation rates. He posted those second thoughts yesterday at: A Short Post on Fixation. My apologies if this is getting boring for Sandwalk readers but I feel an obligation to try and teach creationists about evolution, if for no other reason than being able to say that I tried.

Torley starts with ....

In a previous post, I asked for some experimental evidence to back up Professor Moran’s claim that 22.4 million nearly neutral alleles could have become fixed in the human genome during the last five million years. Were there any other organisms – bacteria, for instance – exhibiting the fixation rate predicted by evolutionary theory for neutral alleles?

Professor Moran kindly provided an example ...

The example I gave was from Richard Lenski's long-term evolution experiment where his group showed the the rate of fixation of neutral alleles corresponded to their mutation rate as predicted.

Initially, I was very impressed with Lenski’s paper, and I was inclined to think that Professor Moran had proved his point. Scientia locuta est, causa finita est. Or so I thought.

A skeptical biochemist

It was then that I was contacted by a scientist who wrote to me, arguing that the fixation of 22.4 million mutations in the human lineage during the last five million years by a combination of selection and genetic drift was impossible and nonsensical for any population of organisms, especially when we consider the pattern of fixation. Strong words! Who was this mysterious scientist? Readers might be surprised to learn that he’s a biochemist with a very impressive track record named Branko Kozulic, whom I introduced to readers in a previous post, titled, The Edge of Evolution? A short summary of his career achievements is available here. Dr. Kozulic also serves on the editorial board of the Intelligent Design journal Bio-Complexity.

By now I was intrigued. Here was a prominent biochemist disagreeing with the arguments of another prominent biochemist! (Larry Moran is a Professor of Biochemistry at the University of Toronto.) Who was right? I decided to investigate the matter further.

Fair enough. Let's see what Branko Kozulic has to say.

Torley begins by pointing out that there is some confusion about the term "mutation rate" and he's right. Sometimes it refers to the number of mutations per round of replication per nucleotide (or base pair). This value is roughly constant and equals about one mutation every 10 billion base pairs (10^-10). Sometimes if refers to the number of mutations per generation or the number of new mutations in every new individual. That value depends on the size of the genome and the number of cell divisions per generation. For humans it's roughly 100 mutations per generation. This can also be expressed as the the number of mutations per base pair per generation. This is called μ (mu) in population genetics.

Theme

MutationIn addition, you sometimes want to talk about the total number of mutations in a population per generation. That value is the mutation rate times the population size. (It's actually the "effective" population size and it only applies, strictly speaking, to populations that aren't growing or shrinking.)

In the experiment I mentioned (Wielgoss et al., 2011), the authors were testing their results against the predictions of population genetics. The prediction is that the rate of substitution of neutral alleles would be equal to the mutation rate of neutral alleles. They determined that there were 941,000 sites in the E. coli genome where a mutation would be neutral. If the mutation rate per replication per base pair was 10^-10 then there would only be 0.0000941 mutations per individual per generation. They sequenced bacteria from the end results of 300,000 generations (multiple cultures at multiple time points). Remember that population genetics predicts that the overall rate of fixation will be equal to the mutation rate per individual per generation (μ).

You would predict 300,000 × 0.0000941 = 28 mutations and the authors found 25 mutations fixed in the population. They conclude that the mutations rate per nucleotide is a bit less than 10^-10. (The paper is a bit confusing on this point since they actually found 35 neutral mutations but some were rejected.)

Thus, they have pretty much confirmed that the rate of fixation of neutral alleles in an evolving population corresponds to the mutation rate of neutral alleles per generation.

Vince Torley and his creationist biochemist aren't convinced. Torley says,

In the passage cited above, Professor Moran referred to Lenski’s long term evolution experiment:

If the fixation rate of neutral alleles was equal to the mutation rate then (as predicted by population genetics) then this should be observable in the experiment run by Lenski (now 60,000 generations).

Did you notice the reference to “the mutation rate”? As we saw above, there are three mutation rates. In chapter two of his book, Population Genetics: A Concise Guide (Johns Hopkins University Press, Baltimore, second edition, 2004), which I’ve been recently perusing, evolutionary biologist John Gillespie repeatedly refers to the mutation rate for a given locus. And in population genetics, altering the numerical relationship between the mutation rate and the (effective) population size can lead to dramatically different results. For example Gillespie, in the textbook referred to above, writes:

If 1/u << N, the time scale of mutation is much less than drift, leading to a population with many unique alleles. If N << 1/u, the time scale of drift is shorter, leading to a population devoid of variation. (2004, p. 31)

Professor Moran is kindly requested to state whether he agrees with this statement, and if not, to provide some references to support his views.

I admit that I should have specifically said "the mutation rate per generation at neutral sites." If that's the important criticism, then I concede that I could have been more clear about what the results were showing.

As for the question, I can't really answer it because I don't know what John Gillespie means by the mutation rate (μ). If it's 10^-10 per bp per generation then 1/μ equals 10¹⁰, or 10 billion. There are very few population sizes that are larger than 10 billion. If Gillespie means something like 10^-4 mutations per generation (E. coli) or 100 mutations per generation (Homo sapiens) then I agree with him, although it seems a bit silly to be talking about a population size of 1/100 (1/μ) in the case of humans.

In the passage cited above, Professor Moran referred to Lenski’s results with E. coli bacteria: a mere 35 fixations after 60,000 generations. That’s about 0.0006 fixation events per generation, for the population as a whole.

By contrast, the fixation rate which Professor Moran claims for human beings (130 per generation) was 200,000 times faster than the rate which Lenski observed for his bacteria. That’s a difference of over five orders of magnitude! This difference in fixation rates requires an explanation. Do we agree on this point, Professor Moran?

I suppose I should have realized that Vincent Torley would not read and/or understand the paper I cited. Maybe I should have explained the result more thoroughly if I want creationists to understand modern evolutionary theory. I hope my explanation in this post helps.

That explains why Vincent Torley had trouble understanding the result. It doesn't explain why "prominent biochemist," Branko Kozulic, who serves on the editorial board of the Intelligent Design journal Bio-Complexity, doesn't understand the scientific literature and evolutionary theory.

Finding the cause that explains the pattern

Now, clearly something was responsible for producing the 22.4 million neutral alleles that distinguish the human lineage from that of chimpanzees. Nobody disputes that. What Dr. Kozulic rejects is the idea that all these mutations could have been fixed by any undirected process (e.g. random mutations plus natural selection, or plus genetic drift), within the time available, especially when we consider the pattern of fixed mutations.

I guess Dr. Kozulic will have to contact me directly and let me know which part of evolutionary theory he doesn't get and which data he rejects.

Isn't it surprising that so many creationists don't understand evolution?

---

Wielgoss, S., Barrick, J. E., Tenaillon, O., Cruveiller, S., Chane-Woon-Ming, B., Médigue, C., Lenski, R. E. and D. Schneider (2011) Mutation rate inferred from synonymous substitutions in a long-term evolution experiment with Escherichia coli. G3: Genes, Genomes, Genetics 1, 183-186. [doi: 10.1534/g3.111.000406]