The "Third Way"/EES and Population Genetics

[swamidass (S. Joshua Swamidass)]

@Mark

Ok. Well, if as Shapiro says, it is true that mutation rates can vary depending on environmental stress (I’m assuming you accept the science here), how are we still able to accurately come up with a mutation rate?

Now that is a totally valid question.

@gbrooks9

You mean for bacteria? Or for higher-level organisms?

This question is valid too, and your answer help us scope this out:

@Mark

Specifically, humans. I don’t know how any of this works. My main question here is: if I accept Shapiro’s Natural Genetic Engineering version of evolution, should I put less trust in the conclusions of population genetics?

Well, to be clear, population genetics woudl not be the problem if we got the mutation rate wrong. The problem would be with the mutation rate. We can put the new rate in, and it merely scales the results up or down by some proportion.

But first, let’s clear some underbrush…

@Mark

Perry Marshall accepts the results of Venema’s work (and I would assume, by extension, Buggs’s and Swamidass’s), but I’m just trying to fit all of this together in my own mind.

That would be odd if he accepts what all of us are saying. Buggs and I disagree with Venema’s work. You can’t agree with all of us simultaneously. Venema wildely overstated the findings of population genetics, and did not even make a sound argument against a bottleneck. So I certainly disagree with his work. Venema’s work is in error, and that became clear in our dialogue on this. Heliocentric Certainty Against a Bottleneck of Two?

---

So, coming back to your question, how do we know the mutation rates? Keep in mind we only need the average mutation rate over time, so it is okay if their is variability. And we have several independent ways of measuring the average rate, and they all produce approximately the same results:

1. Counting the number of mutation differences between chimpanzees and humans, then dividing by the number of years that separate us.

2. Directly measuring the number of mutations between parents and children, after sequencing triads of mothers and fathers and their children.

3. Measuring the amount ancient DNA from our ancestor’s remains differs from modern humans.

All these methods, and more, produce about the same results. We covered this in the dialogue with Buggs. See here:

@swamidass

Directly Measured Mutation Rates

In the genomic age, with easy sequencing, we can now directly measure the mutation rate. Without getting into the details, there are several ways we can do this. More than one method. A recent review, coallates this recent data, and Figure 1 has all we need to assess the mutation rate more carefully. https://www.sciencedirect.com/science/article/pii/S0959437X16301010

image.png1424x1046 140 KB

Each color is a different method, and we can see that they are pretty close, centered at about 0.5e-9. This is a result computed from several different studies, using several different methods, which all give us about the same result. This means that ArgWeaver is using an appropriate mutation rate, determined by several independent methods.

And yes bacteria can mutate much quicker. Buggs raised the issue of the Lenski experiments. And agreed with this analysis I offered…

@swamidass

Variation in Mutation Rates

Yes, there is variation in mutation rate.

However, the complex effects we see, for example, in the Lenski experiment (which was referenced in this conversation, The dynamics of molecular evolution over 60,000 generations | Nature) are not relevant to this problem. Look at figure 2 from the Lenski paper, which shows how mutation rate varies in each experiment. Look at panel B.

nature24287-f2.jpg926x514 82.4 KB

Here we can see there are two groups. One group (red and orange) are mutators , that at some point start rapidly mutating much more than the normal rate. One group (blue - purple) are non-mutators which are just mutating a a more normal rate. The key point is that there is a wild difference between these two groups.

However, this wild of variation in mutation rates is not relevant to mammalian populations . There is much much more constraints on mammalian germline mutation rates, and we do not see such wild swings between populations. So this is an example of an effect in the Lenski experiment that we do not need to account for when studying human DNA. Adding to that pattern, we know that much more of the human genome is non-coding than in bacteria, so it will be more clock like too.

We can measure it in different populations, and we can even detect some differences in the past. These variations, however, in humans are all relatively small. These variations, also, are not always to higher mutation rates, but also to lower mutation rates. So yes, it is likely that mutation rates were slightly higher in particular populations or points in the past (let’s say within 2-fold per year), but it is also likely they were slightly lower at times too. For the most part, this just averages out over long periods when looking at the whole human population. That is not 100% true, but the law of averages is why variation in mutation rate is not going to dramatically increase our confidence interval on TMR4A by much.

Yes, we do expect some variation, but it’s hard to imagine that variation being more than just 10 to 20% when averaged over 10s of thousands of years.

Panel C is important. See how the rate of mutation is a bit higher at first, but then it drops down. Because it is just an initial spike, it just isn’t relevant in the long run. It get’s averaged away by the later lower rates. And these are bacteria too. Mutation rates among humans are much more stable than bacteria. Large mammals have, essentially, a “speed limit” to how many mutations they can endure.

Does that help make sense of this?