More Recent Comments

Saturday, July 09, 2022

Do we need a new theory of evolution?

The classic Modern Synthesis is effectively dead. It was replaced by a more modern version that includes Neutral Theory, Nearly-Neutral Theory, and the importance of random genetic drift. Proponents of the "Extended Evolutionary Synthesis" don't have anything significant to add to our current understanding of evolutionary theory.

The latest kerfuffle in evolution is over a recent article published in The Guardian by Stephen Buranyi, Do we need a new theory of evolution?. The subtitle of the article summarizes the issue ...

A new wave of scientists argues that mainstream evolutionary theory needs an urgent overhaul. Their opponents have dismissed them as misguided careerists – and the conflict may determine the future of biology.

I think Stephen Buranyi did a pretty good job of covering the controversy as long as you ignore the first four paragraphs of his article. He talked to all the right people1 and he got to the gist of the fundamental problem; namely, the over-emphasis on natural selection as the only significant player in evolution. There's no question that this is a serious problem. Here's a quotation from his article.

Doolittle and his allies, such as the computational biologist Arlin Stoltzfus, are descendants of the scientists who challenged the modern synthesis from the late 60s onwards by emphasising the importance of randomness and mutation. The current superstar of this view, known as neutral evolution, is Michael Lynch, a geneticist at the University of Arizona. Lynch is soft-spoken in conversation, but unusually pugnacious in what scientists call “the literature”. His books rail against scientists who accept the status quo and fail to appreciate the rigorous mathematics that undergirds his work. “For the vast majority of biologists, evolution is nothing more than natural selection,” he wrote in 2007. “This blind acceptance […] has led to a lot of sloppy thinking, and is probably the primary reason why evolution is viewed as a soft science by much of society.” (Lynch is also not a fan of the EES. If it were up to him, biology would be even more reductive than the modern synthesists imagined.)

What Lynch has shown, over the past two decades, is that many of the complex ways DNA is organised in our cells probably happened at random. Natural selection has shaped the living world, he argues, but so too has a sort of formless cosmic drifting that can, from time to time, assemble order from chaos. When I spoke to Lynch, he said he would continue to extend his work to as many fields of biology as possible – looking at cells, organs, even whole organisms – to prove that these random processes were universal.

As with so many of the arguments that divide evolutionary biologists today, this comes down to a matter of emphasis. More conservative biologists do not deny that random processes occur, but believe they’re much less important than Doolittle or Lynch think.

You would be hard-pressed to find many popular science writers who have that insight.

So, what's the problem? Why is Jerry Coyne (and others) so upset by the article that he has to write about it twice on his website and send a letter to The Guardian? [Once again: A misguided article on why the theory of evolution is obsolete] It mostly boils down to those unfortunate first four paragraphs. Here they are.

Strange as it sounds, scientists still do not know the answers to some of the most basic questions about how life on Earth evolved. Take eyes, for instance. Where do they come from, exactly? The usual explanation of how we got these stupendously complex organs rests upon the theory of natural selection.

You may recall the gist from school biology lessons. If a creature with poor eyesight happens to produce offspring with slightly better eyesight, thanks to random mutations, then that tiny bit more vision gives them more chance of survival. The longer they survive, the more chance they have to reproduce and pass on the genes that equipped them with slightly better eyesight. Some of their offspring might, in turn, have better eyesight than their parents, making it likelier that they, too, will reproduce. And so on. Generation by generation, over unfathomably long periods of time, tiny advantages add up. Eventually, after a few hundred million years, you have creatures who can see as well as humans, or cats, or owls.

This is the basic story of evolution, as recounted in countless textbooks and pop-science bestsellers. The problem, according to a growing number of scientists, is that it is absurdly crude and misleading.

For one thing, it starts midway through the story, taking for granted the existence of light-sensitive cells, lenses and irises, without explaining where they came from in the first place. Nor does it adequately explain how such delicate and easily disrupted components meshed together to form a single organ. And it isn’t just eyes that the traditional theory struggles with. “The first eye, the first wing, the first placenta. How they emerge. Explaining these is the foundational motivation of evolutionary biology,” says Armin Moczek, a biologist at Indiana University. “And yet, we still do not have a good answer. This classic idea of gradual change, one happy accident at a time, has so far fallen flat.”

I will not defend those statements. They are extremely misleading and they distract from an otherwise decent attempt to explain the controversies in evolutionary theory.

Jerry Coyne, along with Brian Charlesworth and Deborah Charlesworth, wrote a letter to The Guardian in which they defend the old version of the Modern Synthesis with its emphasis on natural selection and fail to mention, except obliquely, the major advance that took place in the late 1960s with the discovery of Neutral Theory [Our letter about evolution to the Guardian (and other stuff)]. He also posted a letter from Douglas Futuyma that wasn't published by The Guardian. Futuyma's letter is much better because he acknowledges the changes in evolutionary theory that took place. Here's what Futyma says,

Mr. Buranyi describes some past challenges to the ES [evolutionary synthesis = Modern Synthesis]. Traditional evolutionary biologists were, indeed, taken aback by evidence that considerable evolution at the DNA level was due not to natural selection, but to random genetic drift – which had already been developed in theory by the geneticists who led the ES. They were taken aback not because of blind faith in the supremacy of natural selection, but because of abundant evidence that even slight differences in organisms’ features were affected by natural selection. Random evolution at the DNA level is certainly the most thoroughgoing change, or expansion, of evolutionary biology since the ES.

I talked to Futuyma about this at the 2015 Royal Society meeting where I suggested that the development of Neutral Theory was about as close to a paradigm shift as you could get. He agreed, but he still prefers to view this as an extension of the Modern Synthesis rather than overthrowing it. Here's what he wrote in the article that was based on his presentation at the meeting.

Thus, evolutionary theory has undergone enormous expansion since the ES, with the neutral theory of molecular evolution its most radical extension [Evolutionary biology today and the call for an extended synthesis].

As far as I'm concerned, that "radical extension" deserves more attention and that's partly what Stephen Buranyi tried to do in his Guardian article. There are far too many scientists who don't understand or appreciate Neutral Theory and the importance of drift and I fear that Jerry Coyne's anger is somewhat misplaced. There are good things about that article that got overshadowed by the first few paragraphs.

The other issue that upsets Coyne and his allies (and me) is the attention focused on proponents of the Extended Evolutionary Synthesis (EES), many of whom are members of The Third Way of Evolution cult. Some of these scientists and philosophers are kooks who don't really understand evolution. Some of the other EES proponents, such as Massimo Pigiucci, seem to be well-meaning, but misguided. Then there's Evelyn Fox Keller who's in a category of her own. There's also the fact that many of them are supported by the Templeton Foundation, a point that wasn't mentioned in Stephen Buranyi's article. However, I think Buranyi does a pretty good job of explaining the controversy.

In 2015, the Royal Society in London agreed to host New Trends in Evolution, a conference at which some of the article’s authors would speak alongside a distinguished lineup of scientists. The aim was to discuss “new interpretations, new questions, a whole new causal structure for biology”, one of the organisers told me. But when the conference was announced, 23 fellows of the Royal Society, Britain’s oldest and most prestigious scientific organisation, wrote a letter of protest to its then president, the Nobel laureate Sir Paul Nurse. “The fact that the society would hold a meeting that gave the public the idea that this stuff is mainstream is disgraceful,” one of the signatories told me. Nurse was surprised by the reaction. “They thought I was giving it too much credibility,” he told me. But, he said: “There’s no harm in discussing things.”

Traditional evolutionary theorists were invited, but few showed up. Nick Barton, recipient of the 2008 Darwin-Wallace medal, evolutionary biology’s highest honour, told me he “decided not to go because it would add more fuel to the strange enterprise”. The influential biologists Brian and Deborah Charlesworth of the University of Edinburgh told me they didn’t attend because they found the premise “irritating”. The evolutionary theorist Jerry Coyne later wrote that the scientists behind the EES were playing “revolutionaries” to advance their own careers. One 2017 paper even suggested some of the theorists behind the EES were part of an “increasing post-truth tendency” within science. The personal attacks and insinuations against the scientists involved were “shocking” and “ugly”, said one scientist, who is nonetheless sceptical of the EES.

Now Jon Perry has entered the fray. Jon Perry is interested in science education and he runs a website called Stated Clearly that emphasizes genetics, evolution, and chemistry. I've posted about his video on evolution where it's clear that he's a proponent of the old-fashioned view of evolution as (mostly) equivalent to natural selection and seems to be a bit confused about mutations and populations [What Exactly Is Evolution? Stated Clearly Gets It Mostly Right]. His latest video (below) is a critique of The Guardian article because it lends credence to those who fight against the "genetic theory of evolution by natural selection." His criticism focuses on those unfortunate first four paragraphs; he doesn't mention the part about random genetic drift being important in evolutionary theory.

This is part of the problem. I don't like the EES proponents but underlying all of that rhetoric is a much more serious problem with the general perception of modern evolutionary theory. There is way too much emphasis on natural selection and it's obvious that this emphasis has permeated the scientific community. It's why we have a controversy over junk DNA.

Jon Perry objects to criticism of the Modern Synthesis, which he describes as the merger of evolution by natural selection with Mendelian genetics. He then touts the Price equation and the mathematical formalization of evolution by natural selection as one of the great achievements of the Modern Synthesis. This is exactly the version that Stephen Jay Gould criticized when he pronounced the death of the Modern Synthesis [Is the Modern Synthesis effectively dead?] Jerry Coyne likes the way Jon Perry "explains" the Modern Synthesis and that's a problem [New video attacks the Guardian’s claim that evolutionary biology is obsolete]. These are complicated issues and I think Coyne and his allies are making a mistake by attacking The Guardian article so vehemently without mentioning the parts that are correct and informative.

Watch the Jon Perry video to see if it really contributes to educating the general public about the current controversies.


1. Full disclosure: I was one of those people.

35 comments :

Bill Martin said...

Hi Larry, I have to ask: Do we have our history right on this? It was Motoo Kimura, Tomoko Ohta and Masatoshi Nei who successfully challenged selection and the New Synthesis by underscoring the importance of randomness (neutrality, Kimura), slightly deleterious mutations (nearly neutral evolution, Ohta) and mutation (Nei). They received quite a few awards and major prizes for that achievement, their accomplishments are no secret. The revolution in evolutionary concepts came from Japan, not from North America as your excerpts tend to suggest. So that is the first thing that needs to be sorted out, one might think. After we have discussed in depth what Kimura, Ohta and Nei achieved in terms of changing evolutionary theory we can discuss what is left in terms of "what's new." If we need a new evolutionary theory (we do not), it needs to be novel in substance, not in label or dialectics. First and foremost it needs a foundational justification. That would start with a list of things for which existing evolutionary theory has no suitable explanation. It is not enough to say "neutrality did it" or "selection did it."

John Harshman said...

Those are major figures, certainly. But so is James Crow, and one should also count all the folks who argued that allozyme variation was too great for selection to account for.

Larry Moran said...

I've always credited Kimura and Ohta (Japan) and Nei ( USA) as well as Jukes, King, Lewontin, Ohno, and others. What surprises me is how little the EES crowd knew about modern population genetics when they first started criticizing evolutionary theory. They also seemed to know nothing about Gould and most of them thought that Dawkins was the leading expert in evolutionary theory.

Joe Felsenstein said...

I think that this is mostly about what name we put on today's theory of evolution. There is always a temptation to declare that we need a brand new theory, which just happens to be the version that the speaker is pushing. I know or knew all of the population geneticists mentioned, and to all of them neutral mutation and random genetic drift are essential processes in evolution. The problem with calls to rename the theory is that it leaves the reader with the impression that the evolutionary processes they learned about in college (selection, mutation, migration, and genetic drift, primarily) have been discredited and discarded. I am not surprised that the EES crowd don't know what is going on in evolutionary biology. What they mostly know is how to call for A New Theory.

Donald Forsdyke said...

THREE HISTORIANS

Yes, journalist Stephen Buranyi did a very good job. Having consulted many presumed authorities, the issue was portrayed as between those who support mainstream evolutionary theory and the EES folk. However, there is a third party that he was not told about. Two distinguished historians (William B. Provine and Mark B. Adams) have made cases over the last three decades that much mainstream theory is wrong. Their cases are supported by "Treasure Your Exceptions." The Science and Life of William Bateson, a second edition of which has just been published in time for Mendel's 200th birthday.

The collective cases of the two historians (WBP & MBA) and an author of the Bateson biography (DRF) are summarized in a new paper recently released in preprint form: "Speciation, natural selection, and networks: three historians versus theoretical population geneticists" This may be obtained from the SSRN preprint server: SSRN Here.

Donald Forsdyke said...

To access the paper, go to to SSRN and enter "Forsdyke".

Steve Gerrard said...

Is neutral drift as important as selection for functional genes? I thought sequence conservation by selection was considered a good indicator of functional genes. How does that work if genes come about by neutral drift?

Arlin Stoltzfus said...

It's Buranyi, not Burani or Burcani. The way that I think about this issue of change is to consider how scientific reasoning has changed. For instance, suppose you see a pattern in which small changes are happening more rapidly to some part of a protein. Under adaptationist reasoning and Fisher's argument, this is adaptation and it involves small changes because those are the most likely to be adaptive. But the neutral theory provides an alternative interpretation in which this part of the protein is the least important. The theory has a mathematical foundation that we can consult to check whether s and N_e are in the right range for our explanation to be plausible. Many other post-Synthesis changes in thinking reflect a recognition of evolution from new mutations, e.g., whenever people are talking today about what is "mutation supply" uN, this is something completely new. The architects of the Mod Synth literally said that the magnitude of the mutation rate was not important because the gene pool was always abundantly full of variation.

William Martin said...

The more I look at this, the more it seems like a non sequitur connecting two questions: (1) Can biologists explain all observable phenomena in the living world at a level of detail, in the geological-historical context, sufficient to convince all dissenters and (2) Do we need a new evolutionary theory because of the obvious answer to (1). The answer to both questions is clearly no. So we should get back to work. Scientists are unable to bring forth a theory that accounts for the origin of life in such a way that has everyone convinced. It will always be that way, whereby we will leave the question of whether that is the fault of "scientists" or "everyone" in that sentence untouched. The same will be true for eyes. The same will be true for eukaryotes. Imagine what the reaction would be if the Guardian were to write "Evolutionary theory can explain everything." We cannot even fully explain the nature of the chemical bond is all detail. So do we throw out quantum chemistry? Or do we get back to work.

Mikkel Rumraket Rasmussen said...

You can never settle a dispute on that question without first coming to agreement on how to measure a degree of importance. I'm serious. What's most important if both contribute to a phenomenon? Well we'd have to find out how to put a unit of measurement on importance. Until that happens anyone can just say they think these four neutral mutations are more, or less, important than that single one with a large selection coefficient.

Marc said...

I think the greatest shift happening in evolutionary biology is on 'the processes by which variations are been generated'.
Contrary to the assumption of SET that variations arise randomly, EES & 3rd way people point out certain internal mechanisms in living organisms that generate variations in a non-random manner.
Concepts such as phenotypic plasticity, non-random rearrangements, epigenetics, transposition, stress-induced mutations and natural genetic engineering add teleology to the generation of variations.
Papers & researchers in support of these ideas are rising every year.

Philosophically, this is a big conceptual shift because evolution driven by internal mechanisms would imply that extant life forms are pre-equipped with the ability to evolve with changing environment.

Arlin said...

Bill, hello! The Mod Synth was said to justify a high-level neo-Darwinian view in which variation merely supplies random raw materials and selection shapes adaptations out of that, which is an all-purpose explanation, i.e., every adaptation can be explained by selection given abundant infinitesimal variation. Allegedly, the Mod Synth made evo bio a real science by giving neo-Darwinism a rigorous pop-gen basis, ultimately reducing it to the action of 4 forces shifting gene freqs in the gene pool, and claiming that all of evo follows from that.

But most people don't understand how the pop gen sausages were made. Squeezing a rationalization of neo-Darwinism out of pop gen requires extreme measures. Nothing in the abstract genetics of pops prevents saltations, extensive neutral evo, mut-limited dynamics, and internal tendencies due to biases in variation, not to mention one-step speciation. The architects of the Mod Synth had to craft a highly specialized version of evo genetics: a polygenic process of shifting gene freqs in a gene pool of abundant infinitesimal variation, where selection is very strong so that it dominates the process of shifting to a new optimum.

That is the basis for the standard view on causation, forces, and things like evo-devo (which is rejected, bc development is not a pop-gen force, just a proximate cause). It's why people in pop-gen have taken ownership of issues of causation.

But everything changes if you start allowing evolution from new mutations, because now the timing and character of evo changes depend on the timing and character of specific muts, which makes it subject to biases in the introduction of new things by mutation and altered development. And this means that mutational and developmental transformations are evolutionary causes in a way that evo-devo people would like to believe. You don't have to be studying gene freqs shifting in a population. That is huge by itself. Now add saltations. Now add neutral evolution.

Certainly many evo biologists today think and talk in ways that are not covered by the old theory, e.g., every time someone mentions genome doublings or symbiogenesis they are breaking neo-Darwinism. Every time someone in pop gen talks about "mutation supply" uN, they are breaking the Mod Synth (the architects literally said that mut rates don't matter bc the gene pool is always full of variation).

But that doesn't mean the old theory has been updated. Theories do not get updated by themselves (I'm sorry: magic does not exist). Mike Lynch and others are still dismissing evo-devo on the grounds of not being about pop-gen forces.

Ultimately we can't fix the Mod Synth by integrating mutationism, macromutations, internal causes of direction, etc., because it was designed to exclude all those things. This is like saying we are going to fix gradualism by adding saltations, or fix the endogenous theory for organelle evolution by adding symbiogenesis.

Larry Moran said...

Thanks for correcting the name. I'm such a poor speller that I run everything through a spellcheck before posting but that didn't help this time.

I was pleased that Buranyi talked to you and the role of mutations got a mention in the article. It's part of the real (not fake) controversy but probably not something that Jerry Coyne and his allies would acknowledge.

I admire Buranyi for taking on a very complex topic and I don't think he is being treated fairly by Coyne.

Larry Moran said...

That view also has religious implications.The fact that life could have been designed to evolve is satisfying to some believers. I'm convinced that this is what motivates some of the Third Way cult members.

Marc said...

Hmm. Whats your position on this random-nonrandom debate?

Arlin said...

We are constantly fighting against this thing where, as soon as we start discussing biases in variation, some people immediately assume that this is *really* about adaptive biases, i.e., they think it somehow implies or requires that the biases are adaptive, or they just want to go down that road and come up with some theory for that. I guess this restores the Darwinian faith (even when mutation biases are in control, selection is really in control making the mut biases), and it is of course fun to think about such things, but from our perspective this is putting the cart before the horse: (1) internal biases such as transition-transversion bias or some developmental biases can be arbitrary, orthogonal to fitness; (2) the theory for the influence of such biases is not classical, not part of the Mod Synth, and depends on new mutations; and (3) the predicted effect has been shown empirically in recent work on mutational effects on adaptations traced to the molecular level.

Mikkel Rumraket Rasmussen said...

"Concepts such as phenotypic plasticity, non-random rearrangements, epigenetics, transposition, stress-induced mutations and natural genetic engineering add teleology to the generation of variations.

No, they don't. How would they? The fact that a bias to some process makes particular outcomes of that process more likely than they would be without that bias (say the probability of deleterious mutations is reduced 3%), doesn't mean the process (or the cause of the bias) is therefore to serve that purpose. At least, if you think it does that view leads you to absurdities, like Earth's gravity is purposefully slamming the moon's backside with rocks from outer space. Whatever you mean by purpose, if you think that qualifies, you and I aren't understanding that word(nor, as a corollary, teleology) to mean the same thing.

Philosophically, this is a big conceptual shift because evolution driven by internal mechanisms would imply that extant life forms are pre-equipped with the ability to evolve with changing environment.

It seems to me evolution driven entirely by mutations subject to natural selection is also "life forms pre-equipped with the ability to evolve with changing environment". That would be true for any process of descent with modification where the modifications(whatever their types and causes) have phenotypic effects that affect their survival and reproduction.

Marc said...

@Arlin
@Mikkel Rumraket Rasmussen

We have multiple experimental observations showing the emergence of mutants with a particular ability from different populations through the same genetic trajectory when all populations were been subjected to the same selective pressure(Taylor et al 2015, Hofwegan et al 2016).
Interestingly, the mutants appeared independently and repeatedly in almost all populations within matter of hours.
These observations indicate that not all mutations are purely random with respect to their origin and survival advantage.

The idea of non-random variations generated by internal biological processes raises a question:
If evolution we see in real time is through these mechanisms, how life forms with this capability/property originated in the first place?

The challenge for the proponents of standard evolutionary theory(SET) is to show that novelties can arise through the fixation of random mutations as well. (overcoming the probability barrier and waiting time problem).

I invite you to list some evolutionary novelties that arose strictly via random/accidental variations.

Anonymous said...

One might also note that phenotypic plasticity is an effect of genotype, not a cause thereof (Baldwin effect excepted), epigenetic inheritance is not stable enough to be relevant to evolution and is unlikely to cross generations in the germ line, stress induced mutations have not been shown to show any different bias than ordinary mutations, and natural genetic engineering is a buzzphrase in search of a meaning.

Mikkel Rumraket Rasmussen said...

@Marc

"These observations indicate that not all mutations are purely random with respect to their origin and survival advantage."
Sure, and in some cases we even understand the mechanisms that cause these specific mutations in response to some environmental cue. And it's true those few and rare regulated mutations can't really be said to be random. But please understand that those types of mutations are extremely few and rare out of all mutations that occur, so their impact on evolution is extremely limited, and is in the few cases where such mechanisms have been found, restricted to switching back and forth between two adaptive responses to environmental conditions often encountered by the organism, like the ability to metabolize two different sugars by some species of bacteria.

"If evolution we see in real time is through these mechanisms, how life forms with this capability/property originated in the first place?"

But it isn't. Those examples are extremely few and rare, and you just can't generalize them to all of evolution.

You will note for example that the vast majority of the genetic differences between humans and chimps are entirely consistent with the known, random biochemical causes of mutations.

"The challenge for the proponents of standard evolutionary theory(SET) is to show that novelties can arise through the fixation of random mutations as well. (overcoming the probability barrier and waiting time problem)."

Ahh I see that you are a creationist. Well that figures.

"I invite you to list some evolutionary novelties that arose strictly via random/accidental variations."
Here you go.

Marc said...

//But it isn't. Those examples are extremely few and rare, and you just can't generalize them to all of evolution//

I disagree.
There are hundreds of studies from the last two decades that reported stress induced non-random variations, epigenetic modifications and phenotypic plasticity.
On the other hand, evolutionary novelties arising through strictly random mutations are limited by probability and waiting time.
As stated by a 2010 Trends in Ecology article,
"In contrast to the rapid response produced by plasticity, if the production of newly favored phenotypes requires new mutations, the waiting time for such mutations can be prohibitively long and the probability of subsequent loss through drift can be high"


//You will note for example that the vast majority of the genetic differences between humans and chimps are entirely consistent with the known random biochemical causes of mutations//


That too I diasgree. Studies report that origin of our species could have involved a large number transposition events.
Other studies have discovered stress induced transposition events, which are of course non-random variations.

References:
(1)https://www.pnas.org/doi/10.1073/pnas.1014330107
(2)https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0131365
(3)https://mobilednajournal.biomedcentral.com/articles/10.1186/s13100-021-00250-2
(4)https://www.nature.com/articles/srep23181
(5)https://www.frontiersin.org/articles/10.3389/fpls.2016.01448/full

//Ahh I see that you are a creationist. Well that figures//
Neither every critics of SET nor every supporters of EES are creationists.

Mikkel Rumraket Rasmussen said...

@Marc


"I disagree.
There are hundreds of studies from the last two decades that reported stress induced non-random variations, epigenetic modifications and phenotypic plasticity."


First of all the fact that the rate of mutation is increased during periods of stress doesn't make those "stress induced" mutations non-random.

It doesn't matter what the total number of studies that discuss some phenomenon is. What matters is what mechanisms the relative proportions of mutations that contribute to adaptation comes from. Out of how many total mutations that contribute to adaptation does adaptation owe to those highly specific, environmentally-regulated, programmed mechanisms of mutation?

So what you need to show is that the majority of adaptive mutations come from those regulated mechanisms that target very specific DNA basepairs for particular types of mutations and reliably induces those specific mutations in response to particular environmental stimuli. And that they don't do this by just inducing mutations all over the place in lots of different individuals and then just have selection work to screen for fitter variants.

Let me save you the time: There is NO study that shows most adaptive mutations come from such mechanisms.


"On the other hand, evolutionary novelties arising through strictly random mutations are limited by probability and waiting time."


Creationist gobbledygook.


"As stated by a 2010 Trends in Ecology article,
"In contrast to the rapid response produced by plasticity, if the production of newly favored phenotypes requires new mutations, the waiting time for such mutations can be prohibitively long and the probability of subsequent loss through drift can be high""


If some imaginary situation obtains then there's no guarantee of evolutionary success. Stop the press.


"//You will note for example that the vast majority of the genetic differences between humans and chimps are entirely consistent with the known random biochemical causes of mutations//

That too I diasgree."


That's not something with which you can rationally disagree. It's an empirically meaured fact. Look at the figures dude. They counted them and characterized the differences by type and their relative proportions are in the figure.

Mikkel Rumraket Rasmussen said...


"Studies report that origin of our species could have involved a large number transposition events."


There isn't anything about what I said above that is in the slightest affected by the observation that transposable elements contribute to evolution.


"Other studies have discovered stress induced transposition events, which are of course non-random variations."


No, they aren't. You just made that up. How did you get to this idea that stress-induced transposition events are non-random? Who convinced you of that and how did they do it? You should contact them and demand and apology for misleading you.


"References:
(1)https://www.pnas.org/doi/10.1073/pnas.1014330107"


Nothing in that paper about the transposons being non-random in any way that implies there is some mechanism that "knows" how or when or where a transposition will be adaptive and transposes adaptively in response without generating a huge number of neutral and deleterious transpositions too.

"
(2)https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0131365"


Nothing in that one either.


"(3)https://mobilednajournal.biomedcentral.com/articles/10.1186/s13100-021-00250-2"

Still nothing there about that.

"
(4)https://www.nature.com/articles/srep23181"


Nor here. This one is about stress-induced transposition by the retrotransposon ONSEN, but ONSEN is like most retrotransposons rather scattershot in it's insertion mechanism(with only a bias towards euchromatin over heterochromatin for rather obvious mechanistic reasons as euchromatin is DNA that is less densely packaged), and so out of 8000 seeds they only found two where insertion resulted in an adaptive phenotype. In other words, it's random transposition and it takes a huge number of them to find an adaptive one.

"(5)https://www.frontiersin.org/articles/10.3389/fpls.2016.01448/full"

This is just a review that discusses the many different ways that transposons can contribute to evolution and gene-regulation. Nothing in this paper shows what fraction of them, much less the majority of transposition events are somehow non-random in any way that implies there is some mechanism that knows how or when or where a transposition will be adaptive and transposes adaptively in response.

So your list of papers supports none of your assertions.

"//Ahh I see that you are a creationist. Well that figures//
Neither every critics of SET nor every supporters of EES are creationists."


I fully concede the capacity of EES proponents to mislead is on-par with creationists.

William Martin said...

Hey Arlin. Let's keep it short. Mike Lynch and others are saying for a decade that everything from the neighbor's dog to the kitchen sink correlates with maximum growth rate (minimum cell division time). Neutral? Drift? What is more selectionist than maximum growth rate? And what is more irrelevant to evolution than maximum growth rate? Larry can guess what is coming now: We hold on to a well known but TRUE observation and inference. E.coli, given enough substrate and 37°C for max growth rate, outweighs the Earth in 48 hours. One bacterium, one species, outweighs Earth. We conclude: Nothing on this planet has ever experienced anything like a maximum growth rate for an evolutionarily relevant amount of time. Oh, but 20 or 20 days or 20 years minutes every 1 million years will do the job. If so then what could possibly be more selectionist, and that from the camp of drift? What, I ask, is the discussion about? One unifying evolutionary ToE from rocks and water to vertical and lateral inheritance to endosymbiosis (SET used in this blog does not mean FRJ Taylor SET to me) to imprinting to complex inherited behaviour (now there is one where all SETs fall short) with only a few neurons. The need for a new evolutionary theory is what folks debate when they forget the dumbfounding observations of nature and worry instead about how good their theory is, which begets a debate of "which theory was that again" and then it is really time to go back to work. Darwin got all the easy stuff right. Leaves the hard stuff. Tough beans. Debating theory, or writing for the Guardian looks a lot easier to me. Its the political Twitter thing: say something outrageous to get atention. Remember the "Darwin was wrong" uproar (The Scientist I think). Yawn. The fix-by-replacement improvement method of badly bruised theories is an interesting procedure. "As we have always said," my axe.









Marc said...

@Mikkel Rumraket Rasmussen

//First of all the fact that the rate of mutation is increased during periods of stress doesn't make those "stress induced" mutations non-random//

Stress induced mutations are non-random with respect to the need of organism because the rise in mutation rate occurs when it is needed i.e during stress.
They are also non-random w.r.t the mode of their origin because its the cell's molecular mechanisms that elevate the rate of mutations when they sense a stressful environment.
I'm not arguing that every individual mutations generated so serve some purpose. My point is: whether the mutations generated do more good or more harm to the organism, they have been generated as a result of organism's problem solving ability, a fact that has been unidentified or underappreciated by SET.
Some studies even suggest that mutations are localised in particular genomic locations.

A 2016 review says:
"Stress-induced mutagenesis encompasses molecular mechanisms of abrupt, transient genomic instability that are upregulated by stress responses, as reviewed here. These mechanisms produce mutations nonrandomly, preferentially when cells are maladapted to their environments—when stressed—and can accelerate adaptation in changing environments. In some mechanisms, the mutations also occur nonrandomly in genomic space: in localized clusters, which could speed concerted evolution within genes, a major problem in the evolution of new protein functions."
Reference:
https://www.annualreviews.org/doi/10.1146/annurev-cancerbio-050216-121919

Go through following two studies. Authors of neither papers mention whether mutations are random or non-random.
But their findings are useful in our discussion.
(1)https://www.science.org/doi/10.1126/science.1259145?url_ver=Z39.88-2003&rfr_id=ori:rid:crossref.org&rfr_dat=cr_pub%20%200pubmed
(2)https://journals.asm.org/doi/10.1128/JB.00831-15?url_ver=Z39.88-2003&rfr_id=ori:rid:crossref.org&rfr_dat=cr_pub%20%200pubmed


//Out of how many total mutations that contribute to adaptation does adaptation owe to those highly specific, environmentally-regulated, programmed mechanisms of mutation?
So what you need to show is that the majority of adaptive mutations come from those regulated mechanisms....//
I only need to show that some of the adaptive mutations are generated by the internal mechanisms of cells. And you concede that (eventhough you argue that its "rare"). In other words, the percentage of non-random mutations doesn't matter. What matters is whether they have an unavoidable role or not.
As I said earlier, the question that the proponents of SET need to answer is: Are randomly generated mutations alone sufficient for the origin of complex adaptations?
(If yes, please provide the supporting emperical data)

You haven't yet replied to that question.


//How did you get to this idea that stress-induced transposition events are non-random?//
Same answer I given for why stress-induced mutations are non-random.

Mikkel Rumraket Rasmussen said...

@Marc


"Stress induced mutations are non-random with respect to the need of organism because the rise in mutation rate occurs when it is needed i.e during stress."


That's a highly idiosyncratic and frankly nonsensical understanding of what it means for mutations to be random, because then all mutations are non-random since they always occur and no organism is perfectly adapted to it's environment. That is to say all organisms are always in need of new mutations because they can always become ever more adapted to their circumstance, and so since mutations keep occurring when they're needed they must all be non-random.

Clearly a ridiculous definition of random is at work then, and a bias in the locations where mutations occur, or an increased rate of them occurring, can't sensibly be said to make them non-random.
Particularly when it is still the case that the vast majority are neutral or deleterious, and to increase the probability that adaptive variants are found, a large number of them must be generated and selection must screen among them.

I mean just think about what that would mean if merely increasing the rate at which adaptive ones occur makes them non-random.

That means if you raise the rate at which adaptive mutations occur from, say, 1 in 5000, to 1 in 2500 cell division, that makes the mutations non-random. Which implies there's some arbitrary rate at which you think it switches over from random to non-random.

So what rate is that? Why are the mutations that occur at the normal–"background" rate of mutations random–and the elevated rate not?


"I'm not arguing that every individual mutations generated so serve some purpose."


But then they are random with respect to their effect on fitness. The mechanisms that result in mutations don't know which ones will be adaptive. They're still produced in blindness to their effects, and success is only increased through the generation of a large number of them that then have to take their phenotypic effects on fitness. All that has changed is that there is an increased rate of occurrence, and a stronger bias in their particular locations.

But again, we have to then ask, if that makes them non-random then where is the threshold where it switches over from random to non-random? How biased in location does it take for random to become non-random? How high must the rate become for it to switch from random to non-random?

How strongly must some change in environment influence these changes in rate, to be considered "teleological" or "purposeful"?

If some chemical pollutant never before encountered by the organism nevertheless has the effect of influencing the rate of mutations in an upwards fashion, and if those mutations that result from this chemical are biased towards particular locations in the genome, is that then a non-random response? Isn't it obvious there must be many chemical compounds and physical forces that will have these effects on the mutation processes simply as a byproduct of whatever mechanical and physical circumstances that happen to be the case. Even something like radiation will unavoidably have such an effect, since at any given moment some DNA will be more tightly wrapped up and protected in nucleosomes, while other parts are expressed, and so that radiation will unavoidably be more likely to produce mutations in some locations, than in others.

Mikkel Rumraket Rasmussen said...


"My point is: whether the mutations generated do more good or more harm to the organism, they have been generated as a result of organism's problem solving ability."


Then the same can be said of any and all mutations that occur. Adaptive ones still occur among them, and they're the result of mechanisms intrinsic to replication, repair, etc. Cell division and replication, and the repairing of damaged DNA is all regulated and adaptive behavior of living organisms, and it all results in mutations, among which adaptive ones occur.

It only makes sense to call them non-random if the specificity and probability of an adaptive one occurring is 100%. A mechanism that reliably responds to a particular environmental cue, targets a specific sequence of DNA bases, and induces one and only one specific type of mutation that is always adaptive under that circumstance(say a particular nonsynonymous mutation in the coding region of the active site of an enzyme that switches it's substrate specificity towards another related sugar that was detected in the environment). Directed mutations of such a specificity and reliability, that and only that, and nothing less than that, can be meaningfully said to constitute a non-random type of mutation.

As soon as it starts being a matter of probability then randomness is now part of the picture, and then it is only a matter of degree of bias in the random process at work. But a bias to a random process doesn't make it non-random, it just makes it biased. If a large diversity of mutations are generated, and if they're spread out over different genomic locations, and all types occur with varying levels of frequency in blindness to their phenotypic effects, then it's random.


"Some studies even suggest that mutations are localised in particular genomic locations.

A 2016 review says:"


Biased towards some locations more than others doesn't make it non-random. Locational bias is of course basically unavoidable since all DNA can't always be equally accessible to physical and chemical influences. Sadly the authors of that paper are just committing the same fundamental mistake you are, by thinking a bias to a fundamentally probabilistic process somehow means it's non-random. For reasons already explained that view simply doesn't make logical sense for it leads to the absurd situation that all mutational processes are non-random as they're all unavoidably and intrinsically biased by whatever local physical conditions apply, and that there is some sort of completely arbitrary change in the rate and locational bias that makes it cross over from random to non-random.

Mikkel Rumraket Rasmussen said...
This comment has been removed by the author.
Mikkel Rumraket Rasmussen said...


"I only need to show that some of the adaptive mutations are generated by the internal mechanisms of cells."


No. All mutations are generated by the internal mechanisms of cells. Some are just more strongly biased than others and occur as a consequence of some environmental stimuli. Whether those internal mechanisms are themselves responses to those stimuli that increase the rate of adaptation doesn't, by that alone, make them non-random.

So no, you really do need to show that the majority of adaptive mutations owe to that highly specific, targeted type of mechanism I described above. Only that can meaningfully be said to be non-random mutations.


"And you concede that (eventhough you argue that its "rare")."


Yes, those highly specific, "directed" types of mutations really are extremely rare and an extremely tiny minority.


"In other words, the percentage of non-random mutations doesn't matter."


Of course it matters if you're going to say that evolution is non-random, whether the majority of adaptive evolution owes to random or non-random effects and processes.


"What matters is whether they have an unavoidable role or not.
As I said earlier, the question that the proponents of SET need to answer is: Are randomly generated mutations alone sufficient for the origin of complex adaptations?
(If yes, please provide the supporting emperical data)"


I'm sorry but it's you who need to show evidence that the majority of adaptive evolution owes to non-random processes of mutation. And no, a mere increase in the bias in the locations at which mutations occur, nor an increase in the rate of adaptation, does not make it non-random.

arlin said...

You guys are talking past each other and being imprecise with language, so getting nowhere fast. I wrote a book about mutation, randomness and evolution. Some day, people will read it and we won't have these crazy discussions anymore.

Experiments and modeling show that sometimes a mutator phenotype has a temporary advantage, under conditions of maladaptation. Therefore, the evolutionary emergence of a regulated mutator state is conceivable and presumably there are some pop-gen conditions under which it is plausible.

But no known system of so-called "stress-induced" mutation has been shown to be an evolved regulated responsive system to improve adaptation by jiggering the mutation rate. Just bc mutation increases under some "stress" does not mean this is regulated and evolved as a regulated response. Marc is jumping to conclusions here. Everything makes the mutation rate either go up or go down. Anything you call "stress" is going to do one of those two things. You can't prove this kind of hypothesis by showing the mutation rate goes up, or even by showing that it goes up in a way that improves adaptation. You have to show that it *evolved* for that reason, otherwise it is just a coincidence.

Meanwhile, there *are* many types of mutation systems that are clearly specialized to probabilize useful mutations. The most well known is the somatic recombination system in antibody generation, which involves programmed DNA rearrangements putting together a V, D and J fusion with junctional shifts and then hypermutation.

That's a somatic system, but there are combinatorial fusion systems in microbes that do similar things, and they are usually involved in immune evasion, e.g., in trypanosomes, borrelia, anaplasma, etc. This is why trypanosomes and borrelia cause relapsing infections: they switch to new antigens using a specialized mut system.

Yeast switches mating types by a mutation that is essentially deterministic.

CRISPR-Cas systems capture foreign DNA and use it in defense. If a phage is not in the environment, the chance that a cell will acquire a defensive spacer is essentially 0. So, the way this happens means that, if we have 3 different environments A, B and C with resident phages pA, pB, and pC, the cell is mostly likely to acquire a spacer against pA in environment A, pB in B, pC in C, and these are also the environments in which the defensive spacer mutation provides the greatest advantage.

So this is a system that induces a correspondence between what is mutationally likely.

Mikkel, the coevolutionary contest between bacteria and phages is quantitatively the most important coevolutionary contest on the planet, involving on the order of perhaps 10^30 deaths per day. So, it is objectively important for evolutionary biologists to study and to capture in their evolutionary theories. And mutation systems that are involved in this context are likewise objectively important, and a failure to cover them is an objectively important failure for a theory.

arlin said...

Sorry, the end of this sentence got cut off: "So this is a system that induces a correspondence between what is mutationally likely...and what is beneficial"

Anonymous said...

What marc is employing is a standard invocation of divine powers when something does not fit a construed formulation of some scientific theory. In this case the construed formulation is that under "evolutionary theory" (whatever that is) all mutation is point mutation and all mutations are equally probable. Science and religion are not alternatives within the same category. Scientists should not engage in such discussions because there is no point trying to convince someone who (i) does not want to be convinced and and (ii) in the realm of beliefs let people believe what they want to believe. Mikkel made a long list of strong points, but no evidence will persuade someone who is convinced that a divine force pushes evolution in such a way to bring forth this or that good or bad trait. By listing evidence, scientists work themselves into a position where marc and his friends can easily find more exceptions, putting the science side on the induced defensive, as if science and religion were alternatives (they are not). Its always the same. Engagement is futile.

Marc said...

@arlin
Glad to hear about your book. Hope I'll buy it someday.

Marc said...

@Mikkel Rumraket Rasmussen

I'd like to know how you distinguish between the mutations in the following two scenarios.

1: An accidental replication error in e-coli

2: e-coli in a stressful condition resulting in elevated mutation rate by the activation and upregulation of error prone polymerase, down-regulation of error correcting mechanisms and movement of mobile genetic elements.

Mikkel Rumraket Rasmussen said...

I've asked you many questions that you've complete ignored about the nature of randomness and what distinguishes it from non-randomness. Please go up and read those questions and give some answers. And in my responses to you I believe I have already provided statements that answer the question you pose to me here.