More Recent Comments

Saturday, February 15, 2014

On the difference between Neutral Theory and random genetic drift

PZ Myers posted an interesting article on The state of modern evolutionary theory may not be what you think it is. He makes the point that there's more to evolution than natural selection.

I think this is an important point but I would not explain it the same way as PZ. He focuses attention on Neutral Theory and the fact that neutral, or nearly neutral, mutations are fixed by random genetic drift. Here's how he describes it ...
First thing you have to know: the revolution is over. Neutral and nearly neutral theory won. The neutral theory states that most of the variation found in evolutionary lineages is a product of random genetic drift. Nearly neutral theory is an expansion of that idea that basically says that even slightly advantageous or deleterious mutations will escape selection — they’ll be overwhelmed by effects dependent on population size. This does not in any way imply that selection is unimportant, but only that most molecular differences will not be a product of adaptive, selective changes.
The debate over adaptationism is a debate over mechanisms of evolution. Random genetic drift is a mechanism of evolution that results in fixation or elimination of alleles independently of natural selection. If there was no such thing as neutral mutations then random genetic drift would still be an important mechanism.

Let's say you have a clearly beneficial mutation with a huge selection coefficient of 0.1 (s = 0.1). Population genetics tells us that the probability of fixation is 2s or, in this case, 20%. That means that the allele will be eliminated from the population 80% of the time. That's random genetic drift. Similarly, some fairly deleterious mutations can sometimes be fixed by random genetic drift.

Random genetic drift is a mechanism of evolution that was discovered and described over 30 years before Neutral Theory came on the scene.

What Neutral Theory tells us is that a huge number of mutations are neutral and there are far more neutral mutations fixed by random genetic drift that there are beneficial mutations fixed by natural selection. The conclusion is inescapable. Random genetic drift is, by far, the dominant mechanism of evolution.

Many people seem to equate Neutral Theory with random genetic drift. They think that random genetic drift is only important when the alleles are neutral (or nearly neutral). Then they use this false equivalency as a way of dismissing random genetic drift because it only deals with "background noise" while natural selection is the mechanism for all the interesting parts of evolution. I think we should work toward correcting this idea by separating the mechanisms of evolution (natural selection, random genetic drift, and others) from the quality of alleles being produced by mutation (beneficial, detrimental, neutral).

The revolution is over and strict Darwinism lost. We now know that random genetic drift is an important mechanism of evolution and there's more to evolution than natural selection. Unfortunately, this blatantly obvious fact is not understood by the vast majority of people and teachers. There are even many scientists who don't understand evolution.


53 comments :

Sean Boyle said...

Did you see my comment on PZ's article giving you and Arlin Stoltzfus a shoutout? His Curious Disconnect series was fantastic, do you have any other guest posts by him?

I love the idea of constructive neutral evolution. It seems that Lenksi's results highlight the importance of it because the Cit+ gene was contingent on several neutral mutations being fixed in the populations that evolved it. You can always tell an IDiot hasn't done their homework when they attack "Darwinism"; Evolutionary theory has "been there done that" and moved on for over half a century now!

Joe Felsenstein said...

I am not sure what has been proven:

1. That genetic drift is responsible for more changes of gene frequency than other evolutionary forces? True, but since these are zillions of little jiggles of gene frequency and some of the them cancel each other out, it is open to question as to whether this make genetic drift the major force in evolution. Whatever "major force" means.

2. That the many adaptations that we see are not the result of natural selection? Gould and Lewontin made the point in their "Spandrels of San Marco" paper that we could not assume that any adaptation we saw was itself directly brought about by natural selection for that purpose. But the high level of adaptation of living systems (a level they have to have, else they could not survive and reproduce) must have been the result of natural selection. Mutational processes alone and/or genetic drift alone just could not have made a bird that flies or a fish that swims, not ever.

3. That if we conclude that, in addition to natural selection, neutral mutation has caused a lot of molecular differences within and between species, then we have rejected the Modern Synthesis? That is just semantics, but it has consequences, and unfortunate ones. Are readers to infer that the lessons they learned in school, telling them that the adaptations of living organisms are the result of natural selection, are now to be discarded?

I think that it would be a tragedy if the promotion of a series of newer and newer evolutionary theories causes people to forget the one important point -- that if there is no natural selection you will not get (or maintain) adaptations.

Yes, let's celebrate the diversity of evolutionary forces, but let's not throw out the baby with the bathwater.

Larry Moran said...

1. That genetic drift is responsible for more changes of gene frequency than other evolutionary forces? True, but since these are zillions of little jiggles of gene frequency and some of the them cancel each other out, it is open to question as to whether this make genetic drift the major force in evolution. Whatever "major force" means.

Setting aside the example I gave where even highly beneficial alleles are lost by drift, the real question is whether a substantial proportion of phenotypic changes are due to drift. I tell my students to look around them in class and observe all the genetic variation in the human population, some of which has become fixed in some populations. Is all that evolution due mostly to natural selection or random genetic drift?

I think that, as a general rule, most of the differences between closely related species are due to drift and not selection. That makes drift a major force in evolution.

2. That the many adaptations that we see are not the result of natural selection?

Nobody is denying the power of natural selection.

3. That if we conclude that, in addition to natural selection, neutral mutation has caused a lot of molecular differences within and between species, then we have rejected the Modern Synthesis?

First, let's stop saying that we're only talking about "molecular differences" when we talk about random genetic drift. That's a bad habit to get into and it it's often used as a way of dismissing the importance of drift and neutral mutations.

The answer to your question is, yes, I think we have to face up to the fact that the derived, hardened, version of the Modern Synthesis did not give due credit to random genetic drift and it completely ignored Neutral Theory. Even though everyone SHOULD be aware of drift and Neutral Theory, the fact is most aren't. The version of the Modern Synthesis that's being taught is misleading and it's probably best to discard it completely in order to get on with 21st century evolution.,

That is just semantics, but it has consequences, and unfortunate ones. Are readers to infer that the lessons they learned in school, telling them that the adaptations of living organisms are the result of natural selection, are now to be discarded?

In some cases, yes. They are going to be told that many of the things they thought were adaptations are only just-so stories and there's another perfectly scientific explanation in the form of random genetic drift. They will never be able to evaluate the scientific evidence for adaptation if they don't even realize that natural selection isn't the only possibility.

Evolutionary psychology will be the first victim but that won't be a bad thing.

I think that it would be a tragedy if the promotion of a series of newer and newer evolutionary theories causes people to forget the one important point -- that if there is no natural selection you will not get (or maintain) adaptations.

We all agree with you on that point. What we're arguing about is whether adaptionism—an almost exclusive focus on adaptation as the only important mechanism of evolution—has been harmful to the understanding of evolution. I am certain that it has.

colnago80 said...

Prof. Moran should note that many of the folks who comment on creationism who are not biologists (e.g. Mano Singham, a physicist, over at freethoughtblogs) refer only to evolution by natural selection. This seems to be the dominant meme in the non biologist scientific community.

Tom Mueller said...

Darwin was the first to recognize the diversity of evolutionary forces! In other words Darwin was not a strict Darwinist as misunderstood by PZ Myers. In other words - it's not always about optimal adaptation.

For example Darwin waxed eloquent on Sexual Selection in his 1871 book The Descent of Man and Selection in Relation to Sex

Georgi Marinov said...

Sexual selection is not a separate evolutionary force

Tom Mueller said...

Hi Georgi - I agree with Joe that we are arguing semantics here...

Larry: [PZ Myers posted an interesting article on the state of modern evolutionary theory may not be what you think it is. He makes the point that there's more to evolution than natural selection.]

My Point (admittedly not at all profound) - Darwin also said as much (and more) in 1871.

Regarding semantics: No can argue that genetic drift is real - no one can argue that natural selection is real - no can argue there is also more to evolution than just natural selection and genetic drift. Now leave it to the number crunchers and pencil sharpeners to argue relative importance.

I don't think any of us are disagreeing with each other. Frankly this topic is a yawner and demonstrates that yet again the tail can wag the dog. Would PZ Myers even have deemed the topic to have been worthy of mention if he (and others) were not so distracted by the background barking of “IDiots”?

... just thinking out loud.

judmarc said...

Would love to see more about the relationship between types of speciation (are speciation and "macroevolution" synonymous?), natural selection, and random genetic drift.

John Harshman said...

Did you mean to say "no one can argue there is also more to evolution than just natural selection and genetic drift" or "one can argue there is also more to evolution than just natural selection and genetic drift"? I can't tell, and the meanings are opposite. The latter is much more sensible, but who knows?

John Harshman said...

No, speciation and macroevolution aren't synonymous. Speciation is often seen as the dividing line between micro and macro. As for your request, have you read Speciation by Coyne & Orr?

John Harshman said...

There has already been an argument on Pharyngula about whether sexual selection is something separate from natural selection. Some people think it is, and I've never been able to understand why. Apparently Kevin Padian is a big advocate for that position. But again I don't understand why.

judmarc said...

No, speciation and macroevolution aren't synonymous. Speciation is often seen as the dividing line between micro and macro.

Confused - I would have thought your second sentence would imply the opposite of your first. In other words, if one knows one is talking about macroevolution when one sees speciation....

Unknown said...

"_Let's say you have a clearly beneficial mutation with a huge selection coefficient of 0.1 (s = 0.1). Population genetics tells us that the probability of fixation is 2s or, in this case, 20%. That means that the allele will be eliminated from the population 80% of the time._"

Looking forward to the icr/aig headline:
Militant atheist biologist moves 80% of the way toward accepting that random mutations cannot produce beneficial characteristics.

Robert Byers said...

why would scientists know this at all? its only a small crowd who think about these things. New conclusions correcting old ones takes time as creationists know.
It will come.

John Harshman said...

The dividing line between A and B isn't necessarily either A or B. And while some people do consider speciation to be macroevolution, not even those people consider it to be all of macroevolution. Red is a color, but nobody thinks "red" is synonymous with "color".

judmarc said...

Got it, thanks.

Joe Felsenstein said...

Larry --

Setting aside the example I gave where even highly beneficial alleles are lost by drift, the real question is whether a substantial proportion of phenotypic changes are due to drift. I tell my students to look around them in class and observe all the genetic variation in the human population, some of which has become fixed in some populations. Is all that evolution due mostly to natural selection or random genetic drift?

I think that, as a general rule, most of the differences between closely related species are due to drift and not selection. That makes drift a major force in evolution.


Of course, there is lots of drift. Of course, since much of the genome is junk, variations there are not the result of natural selection.

But keep in mind that if the population size is N, a selection coefficient much greater than 1/N will be quite effective in the long run. So if N is 1 million, a selection coefficient as low as 0.000001 cannot be ignored. (Yes, in hominoids, N is substantially lower than 1 million, but I'm talking about more typical species).

I think that, this point aside, we generally agree except for the issue of which label to put on theories.

The answer to your question is, yes, I think we have to face up to the fact that the derived, hardened, version of the Modern Synthesis did not give due credit to random genetic drift and it completely ignored Neutral Theory. Even though everyone SHOULD be aware of drift and Neutral Theory, the fact is most aren't. The version of the Modern Synthesis that's being taught is misleading and it's probably best to discard it completely in order to get on with 21st century evolution.,

Who is teaching this derived, hardened form of the Modern Synthesis? When did it arise? The Neutral Theory first appears no earlier than 1966-1968. Before that did George Gaylord Simpson ignore genetic drift? Did Sewall Wright? After it, did John Maynard Smith ignore neutral theory (he wrote papers on it, as it happens)? Does Brian Charlesworth? (He explained the degeneration of Y chromosomes by invoking Muller's Ratchet).

A big problem is that starting in the 1970s-1980s there were a host of young scientists who thought that they had to promote a "new paradigm" of their own or (horrors!) they would only be doing "normal" science. So those of us trying to achieve some clarity and understanding had to continually beat off waves of paradigm-mongerers. (Or perhaps the word should be paradigm-mongers). We continually had to point out that the brand-new Jim-Blotz-ian Synthesis was really wrong, and the brand-new Jane-Smith-ian Synthesis was really old wine in new bottles. This gets very tiring.

And it has one other negative effect. It persuades the public, through the popular science press, that all that old stuff that they heard about how the reason we have such good adaptations is natural selection, all that is now tossed out, and the new theory shows that it is "because of gene duplication", or "because of mutation of developmental regulatory sequences", or "because of a polyploid event". And the fact that natural selection tosses out most such events is not mentioned or appreciated.

Fine, maybe we now have a New Evolutionary Theory. And it will be proven wrong in 5 years and there will be a Newer Evolutionary Theory, and so on indefinitely. But what will the general public make of these endless declarations of the overthrow of existing evolutionary theories? 'Nuf said.

Tom Mueller said...

Hi John... my bad! I meant the sensible version as in no one can dispute that...

Good catch - thanks.

Tom Mueller said...

@ Joe Felsenstein

[But what will the general public make of these endless declarations of the overthrow of existing evolutionary theories? 'Nuf said.]

Bravo!

I think that part of the problem here is that operational models and working definitions are just that - no more.

For example, it would appear that some participants above subscribe to the erroneous notion there is a single agreed upon definition for "species" & "speciation".

I invite everyone to examine Table 1 on page 182 of this excellent book:
http://www.eecs.berkeley.edu/~christos/evol/compevol_files/speciationchapter.pdf

Let’s examine Stephan J Gould’s favorite example – “Gravity”.

Gravity is both fact and theory as correctly elucidated by Gould. Let’s crank it up a notch:

So is Newton's model of gravity correct - or Einstein's? How about Hawking’s”? Which theory is debunked and now out of date?

Trick question: Newton's model works when launching satellites into space - and Einstein's model works when synchronizing those satellites' clocks with our slower clocks back on Earth (darned theory of relativity don’tchya know). Meanwhile, Hawking's equations go a long way to explain the cosmic dance of galaxies.

Meanwhile, Julian Barbour may appears to have the latest new and improved version on the subject: http://en.wikipedia.org/wiki/Julian_Barbour

I think what needs to be understood here is the concept of "theory" as working "model" that can crank out useful answers in deliberately and precisely defined (perhaps even contrived) situations.

So why can't there be more than one working "designation" for "species"?

Why can’t there be more than one operational model for evolution?

The problem is the scientists’ preoccupation with the public lime-light and their tendency to hubris to garner public fame and attention. Didn’t Larry recently criticize Dr. Carl Woese along exactly those lines?

Post script - That said – the moon really does NOT orbit the earth depending on one’s frame of reference. The moon and the Earth together constitute a binary-planet system that together orbit the sun. OK, let's leave Hill Spheres out of this, for the time being, there still is no retrograde motion. Therefore, my point still remains - everything depends on one's frame of reference! (As Galileo, ironically, was the first to understand!)

http://blog.world-mysteries.com/wp-content/uploads/2011/09/Moon_sun_wrong-300x300.jpg

http://blog.world-mysteries.com/wp-content/uploads/2011/09/moon_sun_correct.jpg

Of course, if one were to evoke Einstein’s theory of relativity and change one’s frame of reference to the entire galaxy; the paths of the moon, the earth and the sun would resemble interlocking corkscrew spirals along some non-Euclidian curve in time-space.

Does this mean that Darwin and Newton are both to be thrown into the historical dustbins of science, to be mentioned no more in schools?

As Joe correctly points out - these distinctions are subtle and often lost on laymen who misconstrue “theory” to be “imperfect” truth as opposed to operational models.

Tom Mueller said...

oops - I meant page 180

Larry Moran said...

Joe asks,

Who is teaching this derived, hardened form of the Modern Synthesis?

Professors of Ecology and Evolution at the University of Toronto in the first year evolution course. Also scientists at the Natural History Museum in New York who taught an online course on how to teach evolution. Plus, practically all the high school teachers in America because it's part of the new science "standards."

At last year's evolution conference in Chicago I asked a "random" sample of graduate students to name the main mechanisms of evolution and 10/30 failed to mention random genetic drift.

Joe, I think you're ignoring a very serious problem in the teaching of evolution. I bet half the PI's in my department think that evolution and natural selection are synonyms.

Peter Perry said...

I agree with prof. Moran. There are indeed many people teaching this hardened version of TE, as well as conducting research from a basis on that hardened version. Two examples:

1) when I was a student in Geology, I had a university professor, a Zoologist, that when asked in class by a student if everything we see in macrophenotypes is a product of adaptation, answered that "almost everything" is. He even gave the retarded example that human chins may be the product of sexual selection (you know, chins are great for growing beards, and women love beards). Just-so adaptationist stories are alive and well.

2) there was a recent paper on developmental evolution using cave-dweling organims as an example. The researchers were talking about loss of eyesight as an adaptation. Personaly I fail to see why is adaptation a better explanation of cave-related blindness then inactivation of genes related to eyesight followed by genetic drift, when it's unavoidable that mutations invisible to selection will eventually occur and get fixed by drift, sooner or later. It's invisible to selection if an organism can't process an image with the brain because there is no light or because there is inactivation of genes related to, say, photoreceptors. The energy saved in image processing by the brain should be about the same in both cases, meaning that it is indistinguishable to selection. Why assume it is an adaptation instead of Neutral Evolution? Remember that many of these organims DO HAVE eyes, those eyes also get infected, etc. There is no reason to assume energy savings or protection against infection would be strongly selected for in those cases. But most people assume so just because adaptation is so entrenched, specialy among zoologists. Ask Dawkins or Coyne what they think is the mechanism for the evolution of blindness or albinism in cave-dweling organims.

There are so many examples of groundless assumption of adaptation in Evolution that it's clear many researchers work with the assumption of selection by default for no good reason. This is a consequence of what has and is being tought for decades.

Tom Mueller said...

Larry – I agree that many teachers and many professors are out of their depth when teaching evolution.

The fact remains that you may be exaggerating and guilty of incorrect extrapolation.

Here is the introductory preface in the curriculum framework for AP Biology:

[Although natural selection is usually the major mechanism for evolution, genetic variation in populations can occur through other processes, including mutation, genetic drift, sexual selection and artificial selection. Inbreeding, small population size, nonrandom mating, the absence of migration, and a net lack of mutations can lead to loss of genetic diversity. Human-directed processes such as genetic engineering can also result in new genes and combinations of alleles that confer new phenotypes.]
The document then goes on to detail the importance of genetic drift in evolution and expectations of student responsibility to model genetic drift mathematically and otherwise.

http://media.collegeboard.com/digitalServices/pdf/ap/10b_2727_AP_Biology_CF_WEB_110128.pdf

IB and Pre-IB documents are even more emphatic. Here is the first one I yanked randomly from a google search:

Students are required to
[…Explain that, in addition to natural selection, mutation, genetic drift, gene flow, and sexual selection are also mechanisms that explain the observable fact of evolution.]

http://www.bvsd.org/curriculum/science/High%20School%20Draft%20CEDs/Science%20HS%20Pre%20IB%20Biology%20CED.pdf

OK – I would have phrased that list somewhat differently – there may be some category errors.

But check out any google search on genetic drift in Biology curricula. I could go on

ITMT – Each and every introductory Biology textbook that I am aware of explicitly and emphatically emphasizes the importance of genetic drift as one of many mechanisms in evolution.

I respectfully submit you and PZ Myers may be creating a tempest in a teapot.

Best regards

Joe Felsenstein said...

Larry, Pedro Pereira, and Tom Mueller --

Those are disturbing statistics (though I take some comfort in Larry's statistic that only 10 out of 30 of surveyed gradiate students at the Evolution meetings left genetic drift out of their list of evolutionary forces). Tom's citations from curricula are also reassuring.

I would say that what these show is not that the Modern Synthesis ignores genetic drift, but that a disturbing fraction of people are not teaching the Modern Synthesis. For example, Dr. Pereira, if you do ask Jerry Coyne about blindness in cave fish you might be surprised that he does not only invoke direct natural selection on vision, but includes mutation, genetic drift, and indirect effects of natural selection for other, correlated traits. (I cite his and Allen Orr's book Speciation as support for this prediction).

In fact, correlated effects of natural selection for other traits has been invoked -- long ago -- as a major player in the degeneration of eyes in cave fish. Not just mutation in the absence of selection, though that would act too.

I see some similarity to what you would find if you applied the same method of argument to engineering. Go listen to the lectures on physics in a Mechanical Engineering department. Then you can come away saying that modern physics scandalously ignores Relativity and that we therefore need to throw out the current theory of physics and have a new theory of physics that includes Relativity.

Tom Mueller said...

@ Joe

re: [Go listen to the lectures on physics in a Mechanical Engineering department. Then you can come away saying that modern physics scandalously ignores Relativity and that we therefore need to throw out the current theory of physics and have a new theory of physics that includes Relativity.]

Bravo !!!! bis bis bis...

you and I are on the same wavelength - see my comment below at the very bottom regarding Newton & Einstein.

I find Larry's stats somewhat surprising - and frankly I am inclined to question this anecdotal evidence. Anybody who has been exposed to the Hardy-Weinberg in standard texts should understand that evolution is not all about natural selection.

Tom Mueller said...

@ Pedro

For what it’s worth – I find bearded women a turn-off. I suspect that has something to do with the importance of human sexual dimorphism in sexual selection.

Mind you I remember a documentary on TV describing a small community in India where women with facial hair were actually considered attractive by the local lads - an lo and behold: sexual selection in this neck of the woods worked its evolutionary "magic" as the local girls were remarkably hirsuite.

FTR http://tinyurl.com/pdbs8ov

… and interestingly enough http://tinyurl.com/qzthpg9

So yes - sexual selection is a "separate evolutionary force".

Peter Perry said...

I would say that what these show is not that the Modern Synthesis ignores genetic drift, but that a disturbing fraction of people are not teaching the Modern Synthesis.

Oh, I certainly agree. I'm not making the case that the Modern Synthesis ignores genetic drift or Neutral Evolution (I'm using Modern Synthesis here as not referring strictly to the 1940's view, although drift was known by then). What I'm referring to, and I suspect prof. Moran, is that althiugh textbooks refer to Neutral Evolution and Drift, most teachers I've met tend to pay little more than lip service. When it comes down to their work, they simply "forget" about it and suddenly everything is selection if the case in question is a (macro)phenotype issue.



For example, Dr. Pereira

I'm actualy not a PhD (yet), but thanks. ;)



if you do ask Jerry Coyne about blindness in cave fish you might be surprised that he does not only invoke direct natural selection on vision, but includes mutation, genetic drift, and indirect effects of natural selection for other, correlated traits. (I cite his and Allen Orr's book Speciation as support for this prediction).

In fact, correlated effects of natural selection for other traits has been invoked -- long ago -- as a major player in the degeneration of eyes in cave fish. Not just mutation in the absence of selection, though that would act too.


I was probably a bit mean when referring to Coyne, although I suspect Dawkins would probably go for such a simplistic explanation if asked. Yes, I'm aware the situation is more complicated than the way I portraid it, but it's more or less illustrative of what I hear more often than I'd like to.


I see some similarity to what you would find if you applied the same method of argument to engineering. Go listen to the lectures on physics in a Mechanical Engineering department. Then you can come away saying that modern physics scandalously ignores Relativity and that we therefore need to throw out the current theory of physics and have a new theory of physics that includes Relativity.

I don't know if you had me in mind when you made this specific coment, but as I hope I made clear above, I'm not criticizing the Modern Theory as it stands (although there are certainly aspects open to criticism) but the way it is tought, even in Evolutionary Biology courses, as well as the pervasive assumption that "if you can see it, it was selection", which is far more common than we'd like.


@ Tom Mueller

No one is denying that your particular example makes sense as selection; the point is if it makes sense to assume that human chins were sexually selected for because they make a pretier beard. I strongly doubt that's the case, but it's fine if you believe so.

Tom Mueller said...

@ Pedro - point well taken (although I suspect you never bothered to examine my citations)

I think we can both agree that my citation of chin morphology cogently counts as an example of sexual selection - ditto beards...

I never suggested that chin size was probably selected for in order to grow prettier beards - just that both beards and chin sizes are both evolutionary responses to mechanisms other than natural selection.

Larry Moran said...

Read the Wikipedia article on the Modern Synthesis. I think it fairly represents the consensus view. The emphasis is on natural selection and the role of drift is suspect.

Also read Is the "Modern Synthesis" effectively dead? and leave comments there if you disagree with the consensus understanding of what the "Modern Synthesis" actually means.

Joe is upset about all the modern kinky ideas that are supposed to be overthrowing evolutionary theory and I agree with him that this is silly. However, almost all those "revolutionaries" seem to be completely ignorant of random genetic drift and Neutral Theory. In other words, they missed the real revolution.

Larry Moran said...

Tom Mueller says,

The fact remains that you may be exaggerating and guilty of incorrect extrapolation.

No, Tom, I'm not. I've been studying this for 25 years and there's absolutely no doubt at all about the conclusion. Adaptationism is alive and thriving among the general public and even among university science students. There may be a few good AP Biology teachers out there who are teaching about drift and Neutral Theory but there aren't having much of an impact.

For the past seven years I've been handing out a questionnaire to second year biology students on the first day of class. They all took an evolution course the previous year. Only about 20% of them are able to identify random genetic drift as a mechanism of evolution. When asked to DEFINE evolution, typically 75% of them define it in terms of selection, adaptation, or natural selection.

Why don't you spring a surprise quiz on your own students and ask them to write as sentence of two that DEFINES evolution? I'm willing to bet that most of them will equate evolution with natural selection in spite of what you think you're teaching.

Unknown said...

I agree with Tom Mueller that every single text I've seen that discusses evolution (including human genetics books for nonmajors) prominently addresses drift.

I've also found that questioning undergraduates about evolution (in cell biology and microbiology classes) regularly elicits natural selection as its essence. A little questioning and discussion, though, regularly shows that they have been exposed to drift and seem to grasp it.

I confess that Larry's uh.. emphasis on this issue has caused me to sharpen my focus on it in the above-mentioned classes over the years.

Ben Murray

Tom Mueller said...

Hi Larry – you raise an interesting point of contention.

Very interesting – I think I now understand where you are coming from.

For example, I am exasperated that students can often walk away from 1st year University Genetics courses with an even worse understanding of Genetics than before. I applaud and employ PZ Myers' clarifications such as

http://scienceblogs.com/pharyngula/2011/12/19/steve-pinkers-hair-and-the-mus/

My students understand that Mendel is the exception that proves the rule and not the rule!

Forgive my presumption, I hope my efforts on evolution in class can pass muster. Regarding any pop quiz - here is a SAMPLE of some of the questions (cut and pasted word for word) from my evolution test for Biology 12 students NOT enrolled in AP.

Actually – I have an ulterior motive here. I would love feedback and suggestions for improvement from any and all. I hope it passes muster – remember this is only a selection of questions – not the entire test. (eg – pseudogenes and endogenous retroviral insertions are also discussed)

Test sample from a regular high school evolution test:

Someone tells a Biologist that the “missing link” has never been discovered so Evolution is only just a “theory”! How would a Biologist respond? (hint: there are at least two parts to the answer)

This diagram [of different anteaters on different continents] describes typical observations that Darwin made during his five year voyage on the Beagle. This phenomenon is called: _____________________

Why would these different kinds of anteaters support Darwin’s theory of evolution as opposed to a religious belief called “special creation”? (hint: only one is a marsupial)

What exactly is the “theory of evolution”? Simply that…

List and describe six lines of evidence that support the Theory of Evolution (some have already been mentioned on this test)

Do chromosomal rearrangements guarantee reproductive isolation? Explain with examples…

Can reproductive isolation occur without chromosomal rearrangements? Explain with examples…

[a photo of Ardi together with a diagram of hominid pylogeny is provided] Some scientists have suggested Ardipithecus ramidus (Ardi) is “the missing link” between humans and apes. Others disagree. What is/are the problem(s) here? We have the fossil, so why can’t we all agree? [I am hoping students recognize that speciation cannot be determined by morphology alone and recognizing that “missing link” is a loaded term often pregnant with misconception]

Darwin was the first to concede that Natural Selection was not the only explanation for Evolution

Describe in detail TWO other mechanisms.

Describe two classes of genes whose mutations would allow for very sudden evolutionary changes:

Give an example (of such a sudden evolutionary change) we studied in an earlier unit: (hint – it was on a worksheet in the earlier DNA unit)

Last question to Larry:

Larry - I am but a humble high school Biology teacher. I hope that my former students would do well on any of your pop quizzes.

ITMT – trust me, this is boiler-plate for AP teachers, nothing at all remarkable. Check out the NEW AP curriculum framework again. Hopefully the newer batch of students coming up the the pipeline will be more to your liking.

In any case, I know I am not alone. Check out this site: http://www.indiana.edu/~ensiweb/lessons/ws.extensions.html

Tom Mueller said...

@ Ben

Point well taken - I think I may need to re-examine "emphasis" as well and make sure none of the trees get lost in the forest.

Piotr Gąsiorowski said...

Hmm... Where do they mention mechanisms other than natural selection producing adaptations?

http://www.indiana.edu/~ensiweb/evol.fs.html

Tom Mueller said...

Hi Piotr

I find the ensi site is difficult to search... but a first quick aproximation to answer your question

http://tinyurl.com/kbjowcm

http://tinyurl.com/m5kww2r

Tom Mueller said...

oops and I almost forgot

http://tinyurl.com/jwtjdm6

Piotr Gąsiorowski said...

They have things like "A HARDY-WEINBERG EXCELL SPREADSHEET FOR GENE FREQUENCY CHANGES DUE TO SELECTION" there (bold type added). They do mention the word "drift" but do you think they understand what the Hardy-Weinberg equilibrium is?

Chas Peterson said...

"I think that, as a general rule, most of the [phenotypic] differences between closely related species are due to drift and not selection."

Please at least admit that your "thought" is not supported by any data. And that people who study speciation (Coyne, for one) would, in general, disagree.

Larry Moran said...

Chas Peterson says,

Please at least admit that your "thought" is not supported by any data.

Not a problem ... as long as YOU admit that no other explanation is supported by any data. Deal?

BTW, Jerry Coyne would not disagree.

John Harshman said...

Jerry would agree with this much: that those differences that result in reproductive isolation between closely related species are mostly due to selection. Whehter those particular differences constitute the majority of differences may not be clear, but it's certainly suggestive.

Larry Moran said...

Jerry Coyne believes that most, but not all, of the differences that lead to reproductive isolation are due to natural selection acting on one of the species to make breeding with the other population impossible. There are very few examples that are well-supported. Jerry has published some of them.

I think there are many evolutionary biologists who think that the development of reproductive isolation is usually due to random genetic drfit. I think their view is more likely to be correct.

But, as you note, most "species" are recognized by other phenotypic differences that distinguish them long before reproductive isolation evolves.

SRM said...

Sorry I am not read up on these matters nor deeply familiar with evolution theory, so the following might be too full of misunderstanding for anyone to comment on: but what is the basis for suggesting that reproductive incompatibility is largely due to selection? Surely there would not be direct selection for mutations that lead to a biochemical basis for reproductive incompatibility, would there? Is it rather a case of linkage to other alleles that are subject to selection? Or are we simply talking about the physical tendency or ability to interbreed upon phenotypic divergence?
Otherwise it would seem, regardless of how phenotypically different two groups become upon, say, geographical isolation, that the eventual random fixation of neutral or nearly neutral alleles would be the most likely explanation for a true biochemical inability to reproduce.

Tom Mueller said...

Larry – forgive me, but I am getting too old for this sort of thing. My brain is no longer as agile as when I was younger.

I am wondering out loud if some present (myself definitely included) occasionally fall in the trap of conflating two completely separate notions.
“Random Genetic Events” vs. “Random Genetic Drift”… the two are not equivalent.

I remember enrolling in an Evolutionary Genetics undergraduate course in university. Our prof was most emphatic back then – species and populations within species manifest much variability and the majority of this variability is frankly an accumulation of genetic bric-a-brac – neither useful nor deleterious – his example was hair colour. That was 30+ years ago!

OK – let’s not get bogged down with the relative benefits of differential melanin deposition vis-à-vis Vitamin D deficiency vs. susceptibility to melanoma.

My take home message back then was that the majority of genetic variability was simply unimportant – i.e. neutral

But – and this is the important bit - As soon as formerly unimportant variability does indeed become important, then ipso facto, that very variability becomes subject to Natural Selection and is no longer neutral!

My favorite mammal is the Przewalski’s Horse which provided me the aha moment to finally comprehend exactly how reproductive isolation paces speciation.

SOMETIMES, species that have a different number of chromosomes can’t breed and create viable offspring. For example, domestic horses have 64 pairs of chromosomes and donkeys have 62. When they breed and give birth to a mule, with 63 chromosome pairs, it is usually sterile.

The Przewalski’s horse has 66 chromosomes. When a P-horse and a domestic horse breed, their offspring are born with 65 chromosomes. And get this - those offspring are usually viable.

So are domestic horses and Przewalski’s horses different species? According to most textbooks – the answer is yes.

Things that make you go hmmm… Consider many perfectly healthy populations of house mice, for example, which can be distinguished from each by examining their karyotypes for fused chromosomes? Are different populations of mouse now different species? What about Polar Bears and Grizzly Bears?

Moving on…

So why aren’t P horses and domestic horses reproductively isolated? … maybe it is just too early in the evolutionary game? … but then on the other hand, maybe because reproductive isolation was unnecessary – i.e. unimportant – i.e not subject to Natural Selection.

I notice that a great number of illuminati get very confused on such subtleties. I cite PZ Myers’ errors regarding so-called “chromosome incompatibilities”: http://scienceblogs.com/pharyngula/2013/07/02/the-mfap-hypothesis-for-the-origins-of-homo-sapiens/

Tom Mueller said...

OK OK… where does this leave us so far?

First of all – this discussion really does not have a handle on what exactly is meant by the terms “species” or "speciation".

Second of all – this discussion also needs to define its terms regarding “neutral” vs. ”subject to selection” especially considering disruptive selection ( I hope I am not begging the question)

I like to think of Bmp4 and morphological Variation of beaks in Darwin's Finches as random genetic events

http://www.sciencemag.org/content/305/5689/1462

What is random as well? …any genetic mechanisms for reproductive isolation separating the populations of finches adapted to their particular ecological niches.

At one point – Galapagoes finches were no different than Przewalski’s and domestic horses. However, the game-changer occurred when hybrids suffered any disadvantage. At that point (by tautological definition) Natural Selection kicks in and becomes more important than Genetic Drift.

I explain to my students that actually this is not at all surprising but expected. Consider Ring Species such as Larus Gulls (OK – that may now turn out to be a bad example, but moving on…)

One simplified way to explain the Bateson-Dobzhansky-Muller Model of speciation is along the following. http://en.wikipedia.org/wiki/Bateson-Dobzhansky-Muller_Model

At least two major genetic changes are required to cause hybrid incompatibility and eventual speciation.

A first major genetic change distinguishes population A from population B. Interbreeding between A & B is possible, albeit with reduced viability.
A second major genetic change distinguishes population B from population C. Interbreeding between B & C is possible, albeit with reduced viability.

Interbreeding between A & C is now impossible.

Otherwise speciation would be impossible! Consider, the lone mutation event that distinguishes the first B individual form the A population. With whom will B now breed other than within the original A population? The A population provides the gene pool for the new B variant to spread.

Otherwise, the B lineage would perish.

(aside to Piotr – everything I discuss above is in fact discussed in greater detail on the ENSI site)

I am becoming unclear on what exactly we really are debating here?

I have this nagging suspicion that the hard done by “hard” interpretation of the “modern synthesis” is really just another “strawman”. Can speciation occur by genetic drift – sure why not – check out paradigm textbook examples such as Squirrels on either side of the Grand Canyon

Here is a common site used by high school teachers and university profs.
http://evolution.berkeley.edu/evolibrary/article/_0_0/speciationmodes_01

Please note the IMPORTANT caveat:

[This change might evolve by natural selection OR genetic drift.]

But, as I explained in my opening above, I am easily confused and I could be missing something.

John Harshman said...

If everyone here would just read Coyne & Orr's Speciation, much would be simpler.

Larry: Jerry Coyne believes that most, but not all, of the differences that lead to reproductive isolation are due to natural selection acting on one of the species to make breeding with the other population impossible.

Not quite. First, speciation doesn't require that breeding be impossible, only infrequent and with selection against hybrid offspring. Second, he doesn't think that the differences are selected for the purpose of isolation, but that isolation is a byproduct of selection acting for other reasons, usually in allopatry where there is no possibility of selection for isolation. (Though reinforcement in later sympatry may contribute too, but only after isolation is well on its way.)

I think there are many evolutionary biologists who think that the development of reproductive isolation is usually due to random genetic drfit. I think their view is more likely to be correct.

Why? Coyn & Orr give a good argument to my mind: drift acts much more slowly than selection, requiring both a much longer period of allopatry and the absence of selection during that period.

But, as you note, most "species" are recognized by other phenotypic differences that distinguish them long before reproductive isolation evolves.

I noted no such thing. I don't think it's true, either.

judmarc said...

If everyone here would just read Coyne & Orr's Speciation, much would be simpler.

Would be nice if it came out for Kindle.

John Harshman said...

Hey, isn't 2S a good approximation only when S is very small?

Joe Felsenstein said...

Kimura's 1962 formula, which is damn-near exact, is that the fixation probability for a new mutant present in a single copy in a population of N individuals, is

(1 - exp(-2s)) / (1 - exp( -4Ns))

so you can check this yourself. For example when s = 0.1, and N = 1,000,000, Kimura's probability is 0.18126, while 2s would be 0.2. For s= 0.01, the fixation probability from Kimura's formula in 0.01980133 instead of 0.02.

John Harshman said...

OK, close enough.

Tom Mueller said...

@ Joe - thank you for that explanation. I remain in your debt!

BTW - I am still scratching my head over what I percieve here as a non-debate.

Please correct me if I a wrong, but hasn't PZ Myers merely rediscovered the Luria–Delbrück distribution?

I must be missing something - what?

Anonymous said...

Well, the formula is for diploid species... :-)

(ok, trolling)

Linzel said...

As a high school AP and IB teacher [and future grade 10 life science NGSS teacher], and you probably know this, we have a HUGE set of content and skills we are expected to teach - effectively. It should not be surprising that natural selection has the advantage to be a more easily observable and appreciated aspect of evolution. Drift and neutral theory are more abstract and require more depth of thought, knowledge and effort to TEACH and APPRECIATE [teacher and student]. The amount of time in the IB dedicated to evolution are maybe 10 hours. Often the depth of understanding is relegated to memorization of the ideas rather than application of concept to a distinct situation. I think I'll stop here.....the challenge to teach this WELL is large.

Linzel said...

on another note: My IB Y2s are almost into review and I'm going to have them read this post AND comments [probably the the older post you had on neutral, maybe PZ's] and have them synthesize the argument. Its a great teaching tool. :)