There are many adaptationists who recognize that random genetic drift exists. They will, when pressed, admit that neutral alleles can be fixed in a population. However, these adapationists pften maintain that visible phenotypes cannot be neutral with respect to survivability. Thus all visible phenotypes, with rare exceptions, are adaptations.Several people have expressed this point of view in the comments on Sandwalk but the most prominent proponent is Richard Dawkins. I often use a quotation from The Extended Phenotype to demonstrate how Dawkins thinks about this issue. It comes from a chapter titled Constraints on Perfection. Here's the complete paragraph; I often use just the part that begins "The biochemical controversy ....[Richard Dawkins on Visible Changes and Adaptationism].
I have tried to show that adapatationism can have virtues as well as faults. But this chapter's main purpose is to list and classify constraints on perfection, to list the main reasons why a student of adaptation should proceed with caution. Before coming to my list of six constraints on perfection, I should deal with three others that have been proposed, but which I find less persuasive. Taking first, the modern controversy among biochemical geneticists about "neutral mutations", repeatedly cited in critiques of adaptationism, it is simply irrelevant. If there are neutral mutations in the biochemist's sense, what this means is that any change in polypeptide structure which they induce has no effect on the enzymatic activity of the protein. This means that the neutral mutations will not change the course of embryonic development, will have no phenotypic effect at all, as a whole-organism biologist would understand phenotypic effect. The biochemical controversy over neutralism is concerned with the interesting and important question of whether all gene substitutions have phenotypic effects. The adaptationism controversy is quite different. It is concerned with whether, given that we are dealing with a phenotypic effect big enough to see and ask questions about, we should assume that it is the product of natural selection. The biochemist's 'neutral mutations' are more than neutral. As far as those of us who look at gross morphology, physiology and behaviour are concerned, they are not mutations at all. It was in this spirit that Maynard Smith (1976b) wrote: "I interpret 'rate of evolution' as a rate of adaptive change. In this sense, the substitution of a neutral allele would not constitute evolution ..." If a whole-organism biologist sees a genetically determined difference among phenotypes, he already knows he cannot be dealing with neutrality in the sense of the modern controversy among biochemical geneticists.Natural selection is the only explanation we know for the functional beauty and apparently "designed" complexity of living things. But if there are any changes that have no visible effect—changes that pass right under natural selection's radar—they can accumulate in the gene pool with impunity and may supply just what we need for an evolutionary clock.
Richard Dawkins
The Extended Phenotype (2005)I have discussed this quotation with Richard Dawkins and I am convinced that it fairly represents his viewpoint. The only quibble would be that Dawkins would probably admit of one or two exceptions where neutral alleles might produce a phenotypic effect. In other words, his statement above is perhaps an example of hyperbole but that's how I always read it anyway. Almost all popular science writers make generalizations of this sort and it's not a great crime.
The bottom line is that Dawkins thinks that neutral mutations cannot have an effect on embryonic development; therefore, they cannot result in a visible phenotype. Dawkins believes that almost all visible mutations will have either a beneficial or a detrimental effect on the survivability of an organism and that neutral mutations are a phenomenon that's confined to the molecular level where they may not even count as evolution.
P-ter thinks that I misrepresent Dawkins by quote mining [Larry Moran caught quote mining]. Here's what P-ter says,
This certainly seems to place Dawkins as an "adaptationist", one who thinks that all differences in phenotypes are adaptations. I was a little surprised by this, but the quote seemed clear, and I wasn't going to take the time to find my original.The next lines P-ter is referring to is the beginning of a new paragraph ...
Luckily, another commenter pointed out that The Extended Phenotype is searchable at Google Books [The Extended Phenotype]. And funny, the very next line after Moran stops quoting is possibly relevant:
He might, nevertheless, be dealing with a neutral character in the sense of an earlier controversy (Fisher & Ford 1950; Wright 1951). A genetic difference could show itself at the phenotypic level, yet still be selectively neutral.P-ter then continues with ...
Dawkins goes on to express some skepticism about some arguments for evolution by drift, but he's certainly not an "adaptationist" in the Moran sense.This is a very serious charge. I'm accused of deliberately distorting Dawkins' position by selective quotation. According to P-ter, Dawkins does not believe what he says in the quoted paragraph. (And elswhere, I might add.) According to P-ter Dawkins believes that mutations with a visible phenotype can be neutral. (We're not talking about one or two exceptions here, we're talking about the generality that applies to a significant percentage of mutations.)
I suppose I'm somewhat naive: distorting someone's argument through selective quotation is a classic creationist tactic, and Moran has written a bit about the propaganda techniques used by that crowd. Little did I know his familiarity is not of an entirely academic sort.
[1] As opposed to "pluralists", as he likes to call himself. For someone who (rightfully, in my opinion) is disdainful of "framing" (the view that scientists need to spin their results in order to resonate better with the public), he certainly knows how to frame.
P-ter's evidence of the crime of quote mining is the first two sentences of a paragraph that appears on the bottom of page 32. You can read it for yourself but it seems obvious to me that Dawkins is raising a possible objection to his claim and then dismissing it. Here are the first few (not just two) sentences of that paragraph: I think they convey the correct intent.
He might, nevertheless, be dealing with a neutral character in the sense of an earlier controversy (Fisher & Ford 1950; Wright 1951). A genetic difference could show itself at the phenotypic level, yet still be selectively neutral. But mathematical calculations such as those of Fisher (1930b) and Haldane (1932a) show how unreliable human subjective judgement can be on the "obviously trivial" nature of some biological characters. Haldane, for example, showed that, with plausible assumptions about a typical population, a selection pressure as weak as 1 in a 1000 would take only a few thousand generations to push an initially rare mutation to fixation, a small time by geological standards. It appears that in the controversy referred to above, Wright was misunderstood (see below) ...A careful reading of Dawkins shows that the objection to his claim doesn't stand because people misunderstood Wright. Thus, according to Dawkins, characters that appear to be neutral really aren't.
I maintain that my original characterization of the Dawkins' position is accurate and his words reflect his true beliefs. I resent P-ter's accusation that I deliberately tried to misrepresent Dawkins by quoting that passage.
Incidentally, P-ter puts words in my mouth. I recognize several different kinds of adaptationist. The worst of them are those who think every visible phenotype is an adaptation of some sort but there are many who do not hold this extreme position. It's simply not true that I say every adaptationist must deny the fixation of neutral alleles with a visible phenotype. Some are easier to mock than others, but it's pretty easy to get most of them going whenever I point out that Dawkins is an adaptationist.
50 comments :
Dawkins is an adaptationist allright. He struggles with accepting the fact neutral phenotypic traits can even exist: "triviality" is not obvious to him; and by this he probably meant that good ole haven of irrefutability "we may not have yet discovered its function"
This, even though phenotypic examples of drift, like the evolution of snail shell patterns, are known since the 1920's. Further, though Dawkins admits the influence on evolution of other factors, he finds that the only interesting thing is adaptation by natural selection. It's the anglo, functionalist way. Paley and Darwin agreed on "perfection and complexity of adaptation" as the grand topic to be explained. Dawkins does, too, and considers that natural selection is a sufficient explanation: That's why everything else seems anecdotal to him.
The funniest thing is that natural selection, while always present , is absolutely by no means sufficient to explain adaptation and complexity. We have moved quite past the XIXth century state of the discussion.
Well...some of us have.
Selectively neutral physiological, morphological, and behavioral features have characteristics not shared with genetic neutralism.
1. Phenotypical neutralism is conditional. In fact, both selective and neutral status are conditional. A neutral trait can become non-neutral as circumstances change, and vice-versa.
>Environment-dependent neutralism: abiotic environment, other species (camouflage, toxicity, etc.)
>Conspecific dependent neutralism: female preference, male agonism
>Phenotypic context-dependent: neutral in isolation, functional in particular combinations with other traits
>Frequency-dependent neutralism: high enough frequency become non-neutral (signaling value, perhaps)
>Sex-dependent neutralism: neutral in one sex but not the other, including parental sex-specific genomic imprinting that is beneficial to offspring of one sex but neutral in another
2. Level-divergent neutralism: Neutral with respect for individual fitness but non-neutral with respect to group, species, or clade selection (a possible example: certain mating recognition mechanisms that promote speciation). In addition, a genotype can be non-neutral at the genomic level (ex. meiotic drive) but neutral at the phenotypic level.
3. Phenotypic neutralism allows for wider-ranging exploration of morphospace, randomly walking through adjacent neutral trait variants until chancing upon selectively relevant structures. In that sense, phenotypic neutralism is the advance guard of adaptation. Indeed, I would go so far as to suggest that adaptation through natural selection often depends on the enlistment of phenotypic neutralism.
Tupaia
OT, but it is very confusing for an illiterate layman to follow the differences in treatment of variation. (In the simple or perhaps minimal model of hereditary variation with selection.) Specifically here:
As far as those of us who look at gross morphology, physiology and behaviour are concerned, they are not mutations at all. It was in this spirit that Maynard Smith (1976b) wrote: "I interpret 'rate of evolution' as a rate of adaptive change. In this sense, the substitution of a neutral allele would not constitute evolution ..."
Here Smith illustrates a conflation (a restriction considering the simple model) between evolution and one of its mechanisms. Dawkins words this as a suggestion that variation must be visible for selection, at all times.
Also IIRC at times you hear that variation for some unidentified reason must be 'independent' of selection. But OTOH you hear about theoretical feedback mechanisms such as displaying a trait for sexual selection combining with the preference for it, in a runaway situation.
As I said, confusing. Which is why I feel far more comfortable when I see such lists at Tupaia's exploring means of how variation behaves and are selected visibly against but also invisibly among (for example genetic hitchhiking), either individually or on some other more or less verified level. [Maybe that is a psychological hang up. Do we feel more comfortable with displaying variation or with selecting? Well, at least when you know next to nuthin' the choice :-P is obvious.]
Larry Moran, on June 27th, wrote:
"p-ter says,
Sewall Wright was beating this drum back in the '30s. maybe drift is only beginning to be appreciated now in some parts of the biology world, but evolutionary biology isn't (or shouldn't be) one of them.
I agree that evolutionary biologists like Dawkins and the other adaptationists should have known about random genetic drift. Isn't it amazing that they don't? [my emphasis]"
funny, now you're no longer claiming that Dawkins has never heard of genetic drift.
Since this is such an important discussion in evolutionary biology, it would be great if you (or someone) could invite Dawkins to write a piece directly addressing this precise topic, either on this blog or on his own site.
Even better, maybe he could do an email interview/debate with you?
P-ter says,
funny, now you're no longer claiming that Dawkins has never heard of genetic drift.
Are you one of those people who don't get irony and sarcasm?
Did you honestly believe my position was that Dawkins never heard of random genetic drift?
P-ter, I don't think you're trying very hard to understand this debate.
According to P-ter, Dawkins does not believe what he says in the quoted paragraph.
of course he believes what he says. he's saying a neutral mutation to one group of people is one that does not change the amino acid sequence of a protein. so a genetically-determined phenotypic change cannot be due to a neutral mutation, in that sense
that is, of course, true (to a certain extent--a synonymous change could be relevant via splicing, trna availability, etc). you wanted to make it sound like he's saying no phenotype could possibly be neutral. he's not, and the next paragraph shows the distinctions he is making.
you're right that Dawkins is skeptical of most claims of neutrality. but it seems you wanted to find a quote that sounded a little "sexier". unfortunately, you had to misrepresent him in order to do so.
"He struggles with accepting the fact neutral phenotypic traits can even exist: "triviality" is not obvious to him"
*yawn* Your "fact" is exactly what the argument is about, and--sorry--that "triviality" is obvious to you is not real convincing to the rest of us. This is exactly analogous to the classic IDiot's argument about "design."
"phenotypic examples of drift, like the evolution of snail shell patterns, are known since the 1920's"
If you're going to argue with and about scientists, you really need to get in the habit of sourcing your statements of fact. Please give a few referenced examples of phenotypic examples of drift, starting in the 1920s. The snail shell thing, AFIK, dates from the mid-50s, and, again AFIK, few if any such examples are uncontroversial. The best evidence for drift that I know of consists of differences in the distribution of phenotypic variation between small and large populations--as expected (there's that 2N in the denominator), drift is only shown to be likely in very small populations. Nobody argues that bottlenecks and founder events can not be important phenotypically.
"natural selection, while always present, is absolutely by no means sufficient to explain adaptation and complexity"
And your evidence for that statement is...? Seriously, again you sound exactly like the IDiots over at Dembski's blog. Exactly.
Please. Help us move into the XXIst century. Say something meaningful that doesn't rely on your personal incredulity.
phenotypic examples of drift, like the evolution of snail shell patterns, are known since the 1920's.
Snail shell pattern frequencies are due to drift if one looks at a small scale, but 1. the Europe wide pattern looks like selection; 2. selection by predators is happening locally. This is not the 20's but the 60's and 70's.
OK. I'm giving up responding to Sanders for Lent. Early Lent this year.
There is a well-discussed problem with this whole discussion, and that is the difficulty of distinguishing drift from weakish selection with field data. With only about 60 years in which the question could even be addressed--a trivial amount of time for any evolutionary process--there really isn't much to say other than:
"Phenotypic traits COULD be selectively neutral!"
"Oh yeah? it's unlikely!"
"Nuh-uh!"
"Uh-huh!"
etc. etc. Not really much of a debate.
Torbjörn, I'm sorry but I can't parse much of your comment. You're right that it's confusing, but maybe it doesn't have to be quite so confusing :)
Here Smith illustrates a conflation (a restriction considering the simple model) between evolution and one of its mechanisms.
What do you mean by "illustrates a conflation?"
Also IIRC at times you hear that variation for some unidentified reason must be 'independent' of selection. But OTOH you hear about theoretical feedback mechanisms such as displaying a trait for sexual selection combining with the preference for it, in a runaway situation.
No! No! These are different things. The first one is: variation occurs without respect to need. Ie. the mutations for antibiotic resistance don't occur because the bacteria are exposed to antibiotics. Otherwise you get Lamarckism.
I'm not sure how you get from there to sexual selection, but the feedback mechanisms connect traits, not variation in the trait. They *require* variation to get the feedback going, but the variation assumed in the models is still of the usual, random, naturally occurring kind.
But seriously. How much more mechanistic and historical detail do we gain by pointing at the eye or the mamalina middle ear and saying "selection for eyesight"? "selection for hearing"?
Only when we get into the actual stuff, the natural history documenting the evolution of these complex traits, the role of exaptation stands in your face; it is not at all this directional perfecting of a function, that deep time abstraction that so deeply enamours the adaptationists
Sander "If you think selection is sufficient to explain complex adpataions"
Come on Sanders, your views are far more radical than that noncontroversial - even banal - statement. As Martin Luther exhorted, sin boldly!
"Anglo"? Please. That's reminiscent of Margulis' rhetoric against the Modern Synthesis. Margulis on neo-Darwinism: "a minor twentieth-century religious sect within the sprawling religious persuasion of Anglo-Saxon Biology." Yadda yadda yadda. There's a long history of continental critiques of alleged Anglo-American reductionist functionalism, whether in philosophy, economics, psychology, or biology. (Remember, you can't write "Pangloss" without "Anglo.")
Tupaia
Alex Vargas ("Sanders")is certainly not a very good advertisement for Larry's Eldredgian mantra that paleontologists have something useful to contribute at the "High Table". He's against selection, but what he's for- "Self-organization"? Hopeful monsters?- is never very clear, certainly in his writing and I suspect in his mind.
Dawkins, as this discussion has sufficiently established, does not deny that phenotypically "visible" mutations can be selectively neutral, but there is little question that his default position is to be skeptical of any particular such claim without sufficient data to establish it. As also noted during this discussion, the data are simply not yet sufficient to determine how justified that skepticism (which, it should be noted, is a very respectable position completely within the mainstream of evolutionary biology however much it may annoy Larry) is. In the absence of such data, we get the same old ideological debates and word-chopping, over and over again. Tiresome.
steve labonne: "in the absence of such data, we get the same old ideological debates and word-chopping"
Agreed, which is why I attempted to specify ways in which phenotypic neutralism is distinct from molecular neutralism: conditional, level divergence, morphospace exploration. With those features as well the respective roles of drift and draft/hitchhiking (as Torbjörn Larsson mentioned) in mind, the investigation of phenotypic neutralism can move beyond those earlier debates.
Tupaia
In the absence of such data, we get the same old ideological debates and word-chopping, over and over again. Tiresome.
amen brother!
I am not your regular paleontologist, so dont take me as representative. However it is quite plainly stupid that anyone could think that paleontologists cannot make a contribution and just wipe them of the scenario. But then it is exactly the cahuvinistic frivolity that we are used to hearing from LaBonne and many adaptaionists. To them only the study of selection is wrthwhile. This is what Ortega would call "La barbarie del especialismo".
This is no word game: adaptationists are ideologically bent on bringing it all down to selection. It's pathetic.
Steve LaBonne says,
In the absence of such data, we get the same old ideological debates and word-chopping, over and over again. Tiresome.
Perhaps it's tiresome but it's necessary. As long as we can't conclude that natural selection is all powerful in the realm of visible phenotypes, then the only reasonable scientific position is to maintain an open mind. Those who have already made up their minds that is has to be natural selection seem to be holding an indefensible position. As long as they continue to advocate adaptationism there will be debate from those of us who think there may be other explanations that are being overlooked.
Over the years we've noticed that it's usually adaptationists—those who are sympathetic to Dawkins—who think the debate is trivial and want it to go away. I wonder why?
Yes, but those who have made up their minds that many "visible" mutations are selectively neutral are EQUALLY going well beyond the facts. It is notoriously difficult to get data adequate to distinguish neutrality from weak selection. So more data and less ideology on all sides would be a good thing, no?
windy:
Sorry about my confusing style. I'm afraid it is a sign of my groping for some solid ground here. Biologically layman, biological illiterate, and thus confused. :-)
What do you mean by "illustrates a conflation?"
Smith takes 'rate of evolution' to be rate of adaptive change, which is fine. When he makes the conflation, and notes it, that adaptive change could be all of the evolutionary process: "... in this sense... not constitute evolution". (And the restriction of the simple model would be to say "selection" instead of "hereditary variation with selection". Not that it would work...)
variation occurs without respect to need.
Yes, no teleology is a given, I presumed it was assumed. So you are saying that 'independent' doesn't refer to selection but to "need"? Do you connect selection with "need" somehow, or is it still merely an expression of non-teleology?
the feedback mechanisms connect traits, not variation in the trait.
I see what you mean, my mistake. [Which I feel I have seen or done before. Ouch!]
Sanders wrote:
I am not your regular paleontologist, so dont take me as representative. However it is quite plainly stupid that anyone could think that paleontologists cannot make a contribution
I don't think Steve was saying that *Palaeontologists* cannot make a contribution . . .
Steve LaBonne says,
Yes, but those who have made up their minds that many "visible" mutations are selectively neutral are EQUALLY going well beyond the facts.
There are not very many people in that camp (depending on what you mean by "many"). What you're seeing is a reaction to the adaptationist stance. Some of us argue strongly for an open mind and this often requires convincing people that a non-adaptationist explanation is possible. That's not the same thing as advocating that neutrality is always the case.
True, there are some visible phenotypes that seem so obviously neutral to me that I'm willing to take a position. There are many others where the presumption of neutrality makes a lot of sense and shouldn't be rejected. There are others where natural selection seems like the obvious explanation.
The difference between the pluralists and the adapationists is that the pluralists have different explanations for different characters. They try and pick the mechanism that makes the most sense. The adaptationists, on the other hand, have pretty much already made up their mind before they even examine the evidence.
Speaking of QUOTE MINING, a creationist told me that this quote shows that Darwin is racist. Does this quote really show that?
"At some future period, not very distant as measured by centuries, the civilized races of man will almost certainly exterminate and replace the savage races throughout the world. At the same time the anthropomorphous apes … will no doubt be exterminated. The break between man and his nearest allies will then be wider, for it will intervene between man in a more civilized state, as we may hope, even than the Caucasian, and some ape as low as a baboon, instead of as now between the negro or Australian [aborigine] and the gorilla."
-- Charles Darwin, The Descent of Man, 2nd ed. 1874, pg. 178
by 2007 standards of course darwin was racist.
He was actually not bad by the admittedly abysmal standards of his time. He was a Wedgewood on his mother's side after all, and that family had featured very prominently in the abolitionist movement. Many passages in his writings express his revulsion against slavery.
sven dimilo says,
... as expected (there's that 2N in the denominator), drift is only shown to be likely in very small populations.
Since there are non-experts here I can't let this pass without comment.
It's not true. Random genetic drift occurs in all populations regardless of size. The rate of fixation of alleles by drift is completely independent of population size. It's equal to the rate of mutation in the population. In small populations there are few mutations and some are fixed. In large populations there are many more mutations and many more are fixed.
The confusion arises because for a specified mutation it is more likely to become fixed in a small population than in a large one. But that's not the same thing as saying that alleles don't become fixed in large populations..
Random Genetic Drift
The adaptationists, on the other hand, have pretty much already made up their mind before they even examine the evidence.
if you're saying that some scientists hypothesize adaptations when they see a phenotypic difference, you're right (note that's very different than having made up your mind). Why is this so worthy of your disdain? Most (all?) of them are aware that before claiming to be correct, they have to show some evidence.
in a previous thread, you gave four "textbook examples" of neutral phenotypes in humans: PTC tasting, eye color, tongue rolling, and blood type. To maintain this, you need to explain the patterns of variation at the PTC locus that are characteristic of balancing selection, the fact that the gene controlling eye color (OCA2) has one of the strongest signatures for natural selection in the human genome, and the accelerated rate of change in blood surface proteins. maybe tongue rolling is neutral, once you map the locus we'll have a better idea.
scientists studying these things did not make their minds up without looking at the evidence, you did.
Torbjörn: Yes, no teleology is a given, I presumed it was assumed. So you are saying that 'independent' doesn't refer to selection but to "need"? Do you connect selection with "need" somehow, or is it still merely an expression of non-teleology?
I can't think of another reason someone would say "variation is independent of selection". Since once you look at the situation post-selection, the amount of variation is not independent of selection, because selection (often) culls variation! (another source of confusion here: are they talking about variation, variance or genetic variance...)
The ontology of some objects, especially if they are bound up with the particularities of history, makes them epistemologically less tractable than others. Evolution with its crippled snail’s progress not to mention its dependence on the vagaries of historical and aleatory events is less epistemologically tractable than say molecular objects. Evolution therefore provides much more scope for variations of opinion that, as already been hinted above, may well be informed not just by differences in experience but also differences in ulterior ideology. The suspicions engendered by the ‘hidden variable’ of ideology leads to distrust. Hence, rather than agreeing to differ, malign motives and devious machinations are ascribed followed by impugning of characters.
My favoring of the pluralist view is simply because being a non-expert I don’t have the specialist where-with-all to pare down its wide class of explanatory structures. However, having said that let me be candid and admit that if I think deeply my ‘pluralist’ view of evolution may be informed by a deeper ideology – it does have links with the peculiarities of my own theism. But I won’t go into that here.
On a similar tack: I disfavor ID on a ideological grounds on at least two counts: Firstly - if the ID specialists could prove that so and so ‘design’ is impossible to arrive at via incremental evolutionary assembly (A proof that I hazard is impossible) you then hit an epistemological brick wall. Secondly - ID introduces that most epistemologically intractable ontology; namely a God who works as if by a kind of ‘magic’. In short I admit that I favor conventional evolution over ID as a result of these deeper philosophical considerations/motivations.
When debating differences of opinion ‘feeble mindedness’ and/or ignorance of the respective parties does, of course, have a bearing (after all, we can all suffer from these) but I think it helps to be conscious of the ‘hidden variables’ provided by the Weltanschauung factor before we start impugning characters. I would also like to know if the cut and thrust found in the above discussion differs fundamentally from the debate with the ‘IDiots’.
To suggest that we sit around until more experimental data comes in to help decide these issues is fair minded and scientific but I don’t think humans work that way: they are more inclined to preempt the issue by filling in the blank spaces in a way suggested by their ulterior world view. Lives are too short and human beings too impatient to wait for all the data to come in and even then it is still open to interpretation.
P-ter says,
in a previous thread, you gave four "textbook examples" of neutral phenotypes in humans: PTC tasting, eye color, tongue rolling, and blood type. To maintain this, you need to explain the patterns of variation at the PTC locus that are characteristic of balancing selection, the fact that the gene controlling eye color (OCA2) has one of the strongest signatures for natural selection in the human genome, and the accelerated rate of change in blood surface proteins. maybe tongue rolling is neutral, once you map the locus we'll have a better idea.
scientists studying these things did not make their minds up without looking at the evidence, you did.
Your point is well-taken but it's a little more complicated than you imply. I was searching for examples of possible visible phenotypes that could be neutral with respect to natural selection and gave the four examples you mention. This does not mean that they must be neutral—it means that they are rather obvious cases where neutrality has to be considered.
You responded in typical adaptationist fashion by picking up on any evidence at all that these traits could be selected. As it happens, I don't believe all the papers you quote so I remain skeptical of adaptationist explanations.
I'm quite curious about OCA2 paper you quote as evidence that eye color is selected. I have already discussed that gene in a previous article [Human OCA2 Gene Is Responsible for Eye Color and Skin Color]. The strong signal of selection probably has nothing to do with blue, brown, a green eyes since those alleles are still segregating in the population. Are you suggesting there's evidence that the allele for one type of eye color has a selective advantage? Which one?
I find it very interesting that whenever I suggest the presence of neutral phenotypic alleles in a population there are always a raft of people who challenge every example by presenting an adaptationist explanation. Of course they claim to be only interested in the correct science but it seems to me that they are starting from the premise that neutral phenotypes present a problem that cries out for an adaptationist explanation.
Well, a lot of people have the intuition that a mutation with a detectable phenotypic effect is unlikely to be strictly neutral selectively. A lot of other people, like you, have the contrary intuition. The truth is that there are currently neither theoretical nor empirical grounds adequate to giving a decisive answer as to which set of intuitions is closer to the truth. Where I might sympathize with you is that in certain quarters where the bias is that almost every phenotypic change carries a nonzero selection coefficient, a bit of bias in the opposite direction may be a useful corrective. But best of all is a genuinely open mind, and I'm afraid I don't always detect that in some of your fulminations on this subject.
My bias, for what it's worth, is that most phenotypic changes are probably mildly deleterious, per Ohta. But even if I were a professional in this field I hope I would be sensible enough to admit that this is little more than a bias.
Steve LaBonne says,
Yes, but those who have made up their minds that many "visible" mutations are selectively neutral are EQUALLY going well beyond the facts.
I challenge your claim. I think there's plenty of evidence that "visible" mutations can be neutral. It may not be conclusive evidence but that's not what you said. You implied that I'm just making up these examples without any evidence whatsoever.
Take blood types, for example. The huge variations in allele frequencies between different groups strongly suggests drift and founder effect.
But in any case, you're somewhat missing the point, aren't you? If there was no evidence one way or the other, then the correct pluralist position is that we don't know if it's an adaptation or not. We need to entertain explanations based on selection as well as explanations based on drift. That's really all I'm doing.
It seems like I'm promoting drift over selection because I'm forced into emphasizing drift scenarios in the face of overwhelming adaptationist bias. Please don't confuse that as "driftism."
Well, our comments probably crossed and you'll note that I pretty much agreed with you in advance about that last bit: "Where I might sympathize with you is that in certain quarters where the bias is that almost every phenotypic change carries a nonzero selection coefficient, a bit of bias in the opposite direction may be a useful corrective."
Steve LaBonne,
But best of all is a genuinely open mind, and I'm afraid I don't always detect that in some of your fulminations on this subject.
Steve, I have a bias. I really think the evidence points to an important role for chance and accident in evolution. I think the role of natural selection is very much exaggerated by most people. Unlike Richard Dawkins and other adaptationsts, I think the fixation of neutral alleles is an interesting part of evolution.
Can I ask a question? Are you on record for criticizing the adaptationist position or is it only the pluralists that you challenge?
I've been on the record plenty of times criticizing naive adaptationist stories, generally in the context of the godawful intellectual toxic waste site known as "evolutionary psychology".
Do you allegedly "educated" adaptationists (biologists by training and research)realize that amateurs are quite level with you on your general evolutionary outlook and philosophy? Yes, you argue side by side and seem to be but one.
After all, you are all fans of Dawkins.It seems that's all it takes to be a glad adaptationist.
I thought I'd just warn you, since your amateur buddies will most often be the kind that love evolutionary psychology and selection for just absolutely whatever.
"So more data and less ideology on all sides would be a good thing, no?"
We agree that less ideology would be good, because you guys are simply ignoring data.
1) Non- adaptive traits do exist. It is not "a hypothesis". We find lots of structures that evolve neutrally. While some think it is always possible to play skeptic ,the molecular level of observations provides truly irrefutable examples. Indeed, most genetic change is neutral.
Dawkins downplays neutral genetic changes saying they are not even mutations if they do not affect the phenotype. Yet we know from the data the effect of a mutation can be greatly context-dependent, on the environment and on other mutations. A mutation without an inmediate phenotype cannot be discounted from the evolutionary process. It's THERE, you know. Yet dawkins ideology of genetic reductionism sees the phenotype as if contained in the mutation:It will be expressed anywhere the mutation goes (not in THIS world, Prof Dick)
2) We have all the data to show how exaptation is involved in the origin of complex adaptations, yet adaptationists insist that complex adaptations are only explicable through directional selection (the "perfecting" of a function)
3) A METRIC trait would be great to show how genes accumulate directionally by the action of selection. Mutations can be accumulated by artificial selection to obtain a notorious change in a metric trait. All I want is a FIELD example of precisely that: how natural selection directs organic change in nature. "Functional complexes" are not metric traits. I need something where it is clear what mutations increase and what mutations decrease the trait.
Is an example of a simple metric adaptation too much to ask? I REALLY don't think so!
People who actually test the neutral theory for a living are not nearly as sanguine as you about the possibilities of definitively establishing the neutrality of any given trait. Maybe they know something you don't. I don't get the impression that you know very much about genetics.
Just to address the ABO blood type example (again), sure, geographic distributions of these alleles are clearly suggestive of drift, most likely founder effects in the more dramatic cases (the lack of the B allele in the New World and Australia, and the near lack of the A allele in South America). And it may well be that over much of the modern world these alleles really are neutral.
But if you want to attribute neutrality to blood type, don't we have to ask why there is a persistent polymorphism? Why hasn't drift fixed one of these alleles? Especially since the same polymorphism occurs in chimpanzees and gorillas, suggesting that the polymorphism has been maintained for 8 million years.
To me, that suggests nonrandomness (if I may just-so for a second, perhaps a heterozygote advantage?). In contrast, calling these alleles neutral is just a hunch, even if it "seems obvious."
Your point is well-taken but it's a little more complicated than you imply.
I'm not implying it's simple. it's more complicated than a simple adaptive story, it always is.
you can of course criticize the papers I've cited (I find the evidence for balancing selection on PTC to be not entirely convincing), but those studies should be taken into account when hypothesizing about those traits.
in terms of OCA2, the allele associated with a predisposition to blue eyes is advantageous in European populations (or was for the last couple thousand years). just because variation is segregating doesn't mean it's not adaptive (evolution is of course ongoing).
OCA2 (as you know) is expressed in a number of places besides the eye, so selection probably isn't like *for* blue eyes, per se, but for some other pleiotropic effect (lighter skin, perhaps). I can't think of any way to distinguish the two, but it's an interesting question.
Take blood types, for example. The huge variations in allele frequencies between different groups strongly suggests drift and founder effect.
So you think the MHC region is evolving neutrally? Blood type, like MHC, is involved in immunity--possibly the class of genes *most* likely to be under strong selection (due to the "arms race" with pathogens). blood cell surface proteins have evolved very quickly over evolutionary time (see the paper cited above), and have probably been under very recent selection.
Other loci that show large allele frequency differences between human populations are lactase and loci controlling skin color (the "textbook examples" of recent/ongoing adaptation). Again, frequency differences do not imply selection, but nor do they imply neutrality.
I know enough genetics not to buy Dawkin's naïve version of it.
Steve DiMilo asks,
But if you want to attribute neutrality to blood type, don't we have to ask why there is a persistent polymorphism? Why hasn't drift fixed one of these alleles? Especially since the same polymorphism occurs in chimpanzees and gorillas, suggesting that the polymorphism has been maintained for 8 million years.
I don't think your facts are correct. That may be what's confusing you.
Here's the abstract of a paper by Kermarrec et al. (1999). It corresponds to what I thought was true.
Like humans, non-human primates express the antigens A and B of the ABO histoblood group system. In chimpanzees, only A and O types are found, while the types A, B, AB, and O are found in macaques. The sequences of exons 6 and 7 of two chimpanzee O alleles (Odel and O(x), two macaque species O alleles (rhesus monkey and crab-eating macaque), and sequences of exon 7 of two major chimpanzee A alleles (A1ch and A2ch) were established. The sequences of cDNAs corresponding to the chimpanzee and rhesus monkey O alleles were characterized from exon 1 to 7 and from exon 4 to 7, respectively. A comparison of our results with ABO gene sequences already published by others demonstrates that human and non-human primate O alleles are species-specific and result from independent silencing mutations. These observations reinforce the hypothesis that the maintenance of the ABO gene polymorphism in primates reflects convergent evolution more than transpecies inheritance of ancestor alleles.
Do you have a more recent reference?
Thanks for the reference. It appears that the situation is indeed more complex than I thought. Pointing the Google in this direction I found the following (references linked in text or below):
Chimps have only A and O alleles; gorillas only B. A, B, and O alleles (and therefore AB genotypes), however, are found in several primate species, including the macaques mentioned, baboons, and several New World monkeys.
The consensus seems to be that the A allele is ancestral for primates, B alleles have evolved several times convergently (as noted in the abstract above), and O alleles show up all the time (because O is a loss-of-function null mutation, any missense or nonsense mutation is by default an O allele).
The sole possible exception to the convergence conclusion is the gorilla B allele, which is very similar in coding sequence to the human B (only 2 substitutions functionally differentiate human A & B proteins, and the gorilla B allele has the same 2). Forensic primers for human DNA alleles applied to other species recognize only gorilla B and chimp A. On the other hand, intron patterns suggest a convergent (to humans) origin for gorilla B and (though this doesn't make sense to me) chimp A.
So I must admit that I was probably wrong about the pre-human origin of the human ABO polymorphism...the intron data in particular are hard to argue with. So now we have this: The primate-ancestral A allele has given rise repeatedly, in many species independently, to a B allele. In humans, the O allele has arisen multiple times (again, not surprising for a null recessive), and is now the most abundant allele in the gene pool. In contrast, O alleles are quite rare in other primates.
All that said, there are still good reasons to doubt neutrality and drift as an explanation for the polymorphism.
1) Even leaving chimps out of it, the human A and B alleles seem to have diverged 4.5 to 6 mya, meaning that the polymorphism has been maintained by trans-species inheritance within the genus Homo. My basic point therefore still stands.
2) The O allele, which has zero function, is the most common in modern humans. This makes little sense without some positive selection pressure, since one would expect new nonfunctional O alleles to be eliminated by drift, if not negative selection (as apparently occurs in other primates).
3) The repeated, convergent evolution of B alleles is also contrary to expectations from a neutral/drift model, suggesting balancing selection.
Lots of information on this subject can be found in the last link above (a pdf of a 1997 paper from Mol. Biol. Evol.), and also this post on John Hawks's blog (see the long quote on ABO about halfway down) and here, which includes discussion of possible adaptive functions with several recent references (at the end).
OK...I learned something, but have spent WAY too much time on this. Hope it's helpful and/or interesting to somebody else.
The O allele, which has zero function, is the most common in modern humans. This makes little sense without some positive selection pressure...
Unless the function had recently become superfluous in humans and the non-functional loci were now free to take over by drift? (I agree that the persistence of the other alleles makes this unlikely, but I wouldn't say it makes "little sense", or am I missing something?)
Hope it's helpful and/or interesting to somebody else.
Yes it's interesting, thanks!
above "non-functional loci" should be "non-functional alleles"
"Unless the function had recently become superfluous in humans and the non-functional alleles were now free to take over by drift?"
Yeah, could be. Good point. I was thinking that, since O alleles occur first as mutations to functional A alleles, they are always going to start out as low-frequency alleles and therefore drift is more likely to eliminate them than spread them. But you're right, if all alleles are equivalently neutral there could be a stochastic increase. Must be my darn adaptationist bias rearing its well-designed but ugly head again.
About this postmodern twist that we are all biased, etc.
Equally legitimate viewpoints can indeed exist where we cannot resolve, even if they openly contradict one another, and it is important that scientists understand this clearly.
But not ANY conflict of viewpoints will fit that case. Each case must be studied; if not, it's just a matter of holding a different viewpoint, to submit any topic to bargaining with "the controversy" . There is such a thing as one view being more or less ideological. There is such a thing as demanding that specific claims be backed up with data.
There is a diversity of factors that incide on evolutionary change. Selection is one of them . But some eclipse all other considerations by placing selection in the center. This claim then feeds upon itself to consider selection as a sufficient answer to evolutionary processes.
Is that the correct mindset? Does this broaden or narrow down our way of studying evolution?
Catching up.
windy, if you read this, thanks for your response! It also warrants an answer:
I'm used to operators with non-commutative ordering in physics. Ie for me variation is still independent of selection since the ordering is always fixed. (Growth of variation, selection, growth of variation, et cetera.)
In any case, I do think it comes out to the same thing. But referring to selection is eliminating an undefined middle man. (What is "need"?)
another source of confusion here: are they talking about variation, variance or genetic variance...
After the synthesis I assume it is variation in allele frequencies. This is also what changes during selection.
Post a Comment