Nothing in evolution makes sense except in light of population genetics

zebra.jpgEvolutionary biologist Michael Lynch has published a paper in PNAS this month in which he takes to task scientists in general, and evolutionary biologists in particular, for interpreting “virtually every aspect of biodiversity in adaptive terms.” His language is aggressive: he throws around words like “untenable” and likens the tendency of biologists to dramatize the power of natural selection to the invocation of an intelligent designer. At the end, he writes, “This tone of dissent is not meant to be disrespectful.” Given that he has specifically criticized some of his colleagues, it’s inevitable that some will be offended. But I love this paper, I’m thrilled it’s out there, and I hope it’s making a big splash.

For those interested in my wee little opinion, here’s why:

Theodosius Dobzhansky wrote, “Nothing in biology makes sense except in light of evolution.” Indeed. In general, scientists agree on this. At most biology seminars, in most lecture halls in most universities, the importance of the evolutionary history of the subject under discussion is not contested. It may be overlooked, misunderstood, or understudied, but it is generally not contested. And what about outside the discipline of biology? Same thing. Anthropology and psychology, for example, often invoke evolutionary explanations to describe human experiences. In other words, within academia, evolution is not controversial. But—wait for it—that’s the problem. I’m not about to congratulate modern academic discourse for accepting the scientific establishment of evolutionary biology, even if the topic is controversial in popular culture. Instead, I’m frustrated with and critical of the overuse of evolutionary explanations where they don’t belong.

The problem is that everyone likes evolution. (Everyone except those who loathe it, of course.) The concept of evolution has extraordinary explanatory power, because evolution answers the why question. Natural selection is the evolutionary process that generates adaptations. Therefore, natural selection is the evolutionary process most likely to be invoked to answer any “why?” question of interest. Why did we survive while the Neanderthals went extinct? Because we had a special beneficial mutation. Why isn’t hemochromatosis more rare? Because our hemochromatosis-prone ancestors survived the plagues of Europe. Why do people have religion? Because tool-making was successful. These are just a few examples of overzealous invocations of natural selection that have got me all bent out of shape, and I’ve only been blogging for a few months. But full disclosure, for readers who are not biologists: this criticism is not original. For example, Stephen Jay Gould and Richard Lewontin wrote a scathing diatribe against oversubscription to the “adaptationist programme” back in the 1970s, which I have already wielded in a critical post here. To put it simply, the issue is that it’s not okay to describe the existence of some biological phenomenon as an adaptation, present or past, just because you can dream up a use for it. In their article, Gould and Lewontin describe how and when it’s appropriate to consider the role of natural selection in shaping the subject at hand. And now Michael Lynch has addressed this topic again, nearly 30 years later. The exciting thing is that Lynch’s criticism is specific and incisive, because it delineates the problem in quantifiable, don’t-argue-with-the-numbers terms of population genetics.

butterfly.jpgWhat’s population genetics? Well, that’s part of the problem, too. Population genetics is a quantitative field that examines how gene frequencies change over time, in populations, under different conditions. It’s math. It’s building small, mathematical models of imaginary populations with prescribed parameters as a way of understanding how evolution can work. Since this field has been around for a while, and since some of the most important wet-lab technologies evolutionary biologists use today are recent inventions, it could be said that our understanding of population dynamics, based on population genetics theory, outstrips our ability to test these concepts empirically. So pop gen theory should inform experimental research, but it doesn’t, not always. Not all evolutionary biologists, let alone general biologists, are trained in population genetics. As Lynch writes, “The field of population genetics is technically demanding, and it is well known that most biologists abhor all things mathematical.” Well, maybe, maybe not. But the point is made—population genetics is overlooked, and that’s a big problem, because the very foundations of evolutionary theory are evidenced by this discipline.

Population genetics outlines four forces in evolution: natural selection, genetic drift, mutation and gene flow. Natural selection is a function of organismal fitness; more fit individuals leave more offspring, and the next generation is better adapted to the environment. But the remaining three forces are nonadaptive because they are not a function of fitness. So it follows that the degree to which natural selection has played a role in the evolution of your favorite subject is a function of the power of natural selection relative to the power of the other three forces. And guess what? Population genetics is really, really good at evaluating just this kind of dynamic. Central to Lynch’s argument are the relationships between effective population size, selection and drift. For example, Lynch challenges the claim that the increase in eukaryotic genome complexity over time is a function of natural selection. First, he points out that the larger a gene is, the more vulnerable it is to disruption, since most mutations are deleterious. Therefore, to be established in a population, a modification that makes a gene larger and more complex—with a greater “mutational hazard”—must either have a strong, immediate benefit, or… the population must be so small as to make selection unimportant. Selection is powerful in large populations, but weak in small ones; this may be the most important, and basic, tenet of population genetics. And guess what? Parameter estimates for organisms of increasing complexity show a sharply decreasing relationship to population size. So much so, in fact, that Lynch claims that “the paths open to evolutionary exploration are fundamentally different between unicellular and multicellular species for reasons completely unassociated with organism complexity.” It’s a simple point, well-articulated and defended with baby population genetics theory. Lynch writes about quite a bit more in this paper, but for me this line of reasoning is the most powerful. I’ve definitely been persuaded to schedule a bit more population genetics reading into my own education!

10 Responses to “Nothing in evolution makes sense except in light of population genetics”

  1. thinkevolution.net » Blog Archive » Opening of unnatural history museum Says:

    [...] of natural selection. But that’s a problem in its own right, addressed in my most recent post about Michael Lynch’s recent PNAS paper. Lynch explains that biologists too often attribute [...]

  2. Dan Gaston Says:

    My own personal view is that invoking evolution itself in the explanation isn’t the problem, but invoking adaptation specifically is often overused. Perhaps my own perspective is a bit biased, since I deal with Molecular Evolution and put a great deal of weight on Neutral and Nearly Neutral changes as well as ones that offer a selective advantage. This parallels population genetics where Neutral Drift acts most strongly on Neutral mutations because it removes selection from the equation.

  3. Laelaps Says:

    Great insights. My personal pet peeve when it comes to “adaptation” comes into play when people misuse the term. There’s plenty of talk, especially in popular articles, about organisms “adapting to their environment,” but to me it sounds like the creature in question is thinking “Ok, I live on a lake shore, so I should probably evolve structures to help me make the best of it.” Obviously we know this is not the case, but I tire of hearing the same implication being spat up by professors and popular science writers.

  4. RPM Says:

    Instead, I’m frustrated with and critical of the overuse of evolutionary explanations where they don’t belong.

    Along the same lines as what the previous two commenters wrote: it’s not evolutionary explanations that are a problem, but adaptationist explanations. Lynch also published a version of this thesis in MBE about a year ago, and I’m pretty sure he also brings it up in his new book (which I have not read).

    I’m not sure if this was intentional or not, but your blog post title is actually Lynch’s motto for his lab’s research. He includes it in his talks and I think it’s also on his lab webpage (did he include it in the article too?).

    Also, you need to post a sitefeed for you blog. I’ve visited a couple of times via various links, and I always look for a feed to add to my newsreader . . . but I can’t find one!!

  5. ABP Says:

    I’m not sure if this was intentional or not, but your blog post title is actually Lynch’s motto for his lab’s research. He includes it in his talks and I think it’s also on his lab webpage (did he include it in the article too?).

    Indeed, totally intentional. Lynch uses this phrase as a section heading in the paper. And just a few minutes ago my copy of The Origins of Genome Architecture, Lynch’s new book, was delivered to the lab. When I get around to reading it hopefully I’ll be able to post some more thoughts on this.

    Thanks much for the heads up on a sitefeed. I still have no idea what I’m doing with this blogging stuff but adding a sitefeed sounds like a straightforward task that I should be able to handle. If any of you other grad student evolutionary bloggers visit this thread again:

    Anyone going to the European evolution meeting in August? That’s my next meeting, would be nice to meet some of you all…

  6. Chris Harrison Says:

    Very nice post. I read Lynch’s paper last night, and I think you did a great job with your summarization. I’m a first time visitor, but I don’t need a sitefeed to remember to come back!

  7. CCP Says:

    We who study organisms as opposed to genes, molecules, or mathematical models do not doubt the reality of adaptation. The concept may well be overused, or used in poorly supported contexts, but please let’s keep the baby.

  8. DGS Says:

    CCP writes: We who study organisms as opposed to genes, molecules, or mathematical models do not doubt the reality of adaptation. The concept may well be overused, or used in poorly supported contexts, but please let’s keep the baby.

    Lynch is an old-school organismal biologist. Reems and reems of Daphnia studies, and his lab hosts the Daphnia stock center. Lynch doesn’t doubt the reality of natural selection, if you read any of his work (the quant gen book, for example) he gives plenty of machinery for studying just that. It’s just that there are four forces at work, and he focuses on the other three.

  9. Shoa Clarke Says:

    I have read the paper and I’ve heard Michael Lynch talk on the subject. Indeed, he is a bit “scathing.” While he brings up some very poignant and important points, he is a bit self righteous, and at times, he just seems like he’s starving for attention. In the paper, Lynch presents a number of “myths” which he dispels. While some of the myths are clearly blatant falsehoods (“Evolution is natural selection”) which certainly should be set straight, others are not really “myths” but simply ideas (“Natural selection promotes the ability to evolve”) which have neither been proven nor disproven. Moreover, in the process of presenting these myths which he claims we (other biologists, the public, etc) should question, he presents some of his own ideas, which are often equally questionable.

  10. Jason Statham Workout Routine Says:

    Jason Statham Workout Routine…

    [...]thinkevolution.net » Blog Archive » Nothing in evolution makes sense except in light of population genetics[...]…

Leave a Reply

You must be logged in to post a comment.