Has the appearance of Europeans lightened up in 5000 years?

[chadn737]

I have cited no studies showing selection coefficients at all, as far as I noticed.

 

I simply observed that skin colors in human populations have evolved rapidly quite often in the past, the genetics are plastic (set up to allow sensitivity to selection pressure, unlike, say, design height or digestive tract physiology), and a couple of selection pressures on the humans populating inland and northern Europe during the post-glacial warming are both obvious and in apparent likelihood very strong. I find that situation unexceptional, actually, among the population expansions of mammals and birds and probably vertebrates generally, but whether it is unique to humans or not it is observation from which the ordinary conclusions (right or wrong) might be drawn without much shock - why not?

 

1) If you don't know what the selection coefficients are for these other populations then statements like this:

 

"I simply observed that skin colors in human populations have evolved rapidly quite often in the past"

 

are unsupported. You don't know that skin color evolved rapidly in these other populations. The rate of evolution is a measurable quantity, not one that you simply claim to be so. Because multiple factors play into this, you don't know that it evolved rapidly. For example, lets say that there is population discontinuity because one group of a different color migrated in and basically wiped out the previous inhabitant and replaced them. This is exactly what happened in North America and Australia. In a handful of generations you have nearly complete replacement of one population by another. If you do not account for possibilities like that, you would erroneously conclude that there was rapid evolution. Hence my continued harping on the importance of population continuity/discontinuity. You are going around and making unsupported claims about human evolution and then saying that these results are not unique...you can't do that overtone, your argument needs to be supported by actual research, in this case you need those selection coefficients, you need the evidence of population continuity or discontinuity.

 

2) This: "the genetics are plastic (set up to allow sensitivity to selection pressure, unlike, say, design height or digestive tract physiology)" is nonsense. How are the genetics "set up allow sensitivity to selection pressure"? What is the mechanism of such sensitivity? These traits are under the same rate of mutation. Scattered around the genome, there is unlikely to be a unusual recombination for these specific loci. The way you are describing this simply makes no sense from the standpoint of genetics, but instead, seems to imply an almost Lamarkian sense of evolution. If this claim made any scientific sense, we could actually discuss, but as it stands it sounds like hand waving nonsense. What determines response to selection is the phenotypic trait, because it is there that selection operates and that is determined ultimately by environment and external factors, not some vague "sensitivity" at the genetic level.

 

And yet another 50 second google, this time to an ordinary Wiki page: http://en.wikipedia.org/wiki/Genetics

 

 

 

This and the subsequent quote from wikipedia:

 

 

 

Genetics is the process of trait inheritance from parents to offspring, including the molecular structure and function of genes, gene behavior in the context of a cell ororganism (e.g. dominance and epigenetics), gene distribution, and variation and change in populations. Given that genes are universal to living organisms, genetics can be applied to the study of all living systems, including bacteria, plants, animals, and humans. The observation that living things inherit traits from their parents has been used since prehistoric times to improve crop plants and animals through selective breeding.^{[6] - - -}

 

have absolutely nothing to do with your claim of genetic plasticity. Explain to me how quoting a textbook definition of genetics explains why you referred to "plasticity" in the context of phenotype? It doesn't.

 

Again, the recommendation would be to give up on that entire approach to my posts. You don't know what you're doing. Just read the words for their meanings in the English language, derive from that reading the meaning of the post, and respond accordingly.

 

 

I'm about to give up trying to make sense of your posts, if thats what you mean. You clearly wrote about plasticity in reference to phenotype. Every time you used the term "plastic" in your original posts, you were referring either to the word "trait" (phenotype) or specific phenotypes (skin tone, vitamin D production, tanning....). You are trying to change your argument midstream and claim that you meant something other than what you clearly wrote.

[overtone]

1) If you don't know what the selection coefficients are for these other populations then statements like this:

 

"I simply observed that skin colors in human populations have evolved rapidly quite often in the past"

 

are unsupported.

They are supported by the observation of 1) numerous differences in genetically controlled and inherited aspects of skin color between geographically separated populations with common ancestry, and 2) the maximum available amount of time the populations involved can, or are likely to, have been separated.

 

The actual selection coefficients imposed would of course depend on the actual intervals over which the changes occurred, the population sizes involved, the plasticity (likelihood of viable alteration) of the genetics involved, etc. They are unknown to me, and irrelevant to my argument.

 

That isn't a deep, difficult, vague, or subtle argument. You are being willfully obtuse.

 

 

 

Explain to me how quoting a textbook definition of genetics explains why you referred to "plasticity" in the context of phenotype? It doesn't.

I was explaining to you why insisting that the word "trait" in other people's posts refers to phenotype only and excludes reference to genetic heritage was and is an error on your part. Note the repeated and ordinary use of "trait" in the Wikiquote definition of "genetics", as something inherited, genetic, etc. It's not just me: lots of people use the word "trait" to refer to something inherited, when discussing genetics, inheritance, evolutionary change, etc.

 

Aside from some attempt to discuss a relevant phenotypic plasticity as an inherited trait (i.e. genetically controlled, evolutionarily alterable), which they occasionally are and definitely in the particular case of the human ability to tan in response to sun exposure (as I noted), you have no reason to insert the term into my posting. Since you didn't do that, you have obviously confused yourself - as recommended, take it out and start over.

 

 

 

I'm about to give up trying to make sense of your posts,

If that is followed by not replying to what you haven't made sense of, rather than re-imagining it to suit yourself, we'll be money and time ahead.

Edited March 18, 2014 by overtone

[chadn737]

They are supported by the observation of 1) numerous differences in genetically controlled and inherited aspects of skin color between geographically separated populations with common ancestry, and 2) the maximum available amount of time the populations involved can, or are likely to, have been separated.

 

The actual selection coefficients imposed would of course depend on the actual intervals over which the changes occurred, the population sizes involved, the plasticity (likelihood of viable alteration) of the genetics involved, etc. They are unknown to me, and irrelevant to my argument.

 

That isn't a deep, difficult, vague, or subtle argument. You are being willfully obtuse.

 

Neither one of those assumptions are sufficient to conclude that a trait has been selected for or rapidly evolved under selective pressure. You have to be able to distinguish actual selection for a trait from demographics. Consider what has happened in the African American population. The Average African American has 10% or more of their ancestry from white Europeans due to slave owners back in the day taking advantage of them. If we were to sample from this population the skin tone of modern African Americans and the skin tone of ancestral African populations, you would likely find some differences. If you were to compare skin tone of modern African Americans to the 10% or so European ancestry, you would find bigger differences. If you look at this naively, you would wrongly and foolishly conclude that there has been rapid evolution due to some environmental selection. When in fact, its all a matter of demographic history, not actual natural selection.

 

I'm not even going to get into how badly you are butchering terminology here, but the fact that all these factors are "unknown to me, and irrelevant to my argument" is disgraceful. If you do not know these factors, then you really know nothing about how something evolved or how rapidly it evolved. Unless you actually test your assumptions, test for actual selection in the genome, prove population continuity.....you don't know whether or not a trait was actually selected for or against, you don't know if it evolved due to natural selection or if it is merely fluke of demographic shifts.

 

I find it disturbing that you constantly avoid trying to support your arguments.

 

 

I was explaining to you why insisting that the word "trait" in other people's posts refers to phenotype only and excludes reference to genetic heritage was and is an error on your part. Note the repeated and ordinary use of "trait" in the Wikiquote definition of "genetics", as something inherited, genetic, etc. It's not just me: lots of people use the word "trait" to refer to something inherited, when discussing genetics, inheritance, evolutionary change, etc.

Aside from some attempt to discuss a relevant phenotypic plasticity as an inherited trait (i.e. genetically controlled, evolutionarily alterable), which they occasionally are and definitely in the particular case of the human ability to tan in response to sun exposure (as I noted), you have no reason to insert the term into my posting. Since you didn't do that, you have obviously confused yourself - as recommended, take it out and start over.

 

Trait does refer to phenotype. Even in the wikipedia article it is referring to the phenotypes. You used it in your posts to refer to phenotypes and when challenged on it, you chose to change your argument rather than simply admit you were wrong. If you truly meant it in some other fashion, then you need to educate yourself in the correct terminology before talking about it.

[overtone]

 

 

Neither one of those assumptions are sufficient to conclude that a trait has been selected for or rapidly evolved under selective pressure.

They are not assumptions, but observations. And while of course insufficient for final and incontrovertible "conclusion", those observations are strong evidence - one would not be surprised to discover that the indicated evolution under selection pressure had in fact happened in one, two, three, four, or even all of the events. And this latest one of course fits that pattern.

 

In fact, looked at from another perspective, the circumstances of the latest discovery might be reasonably taken as supporting evidence for similar hypotheses put forward to explain the many other less easily investigated skin color transformations in humans we encounter around the planet, all within a couple of thousand generations.

 

 

 

You have to be able to distinguish actual selection for a trait from demographics. Consider what has happened in the African American population.

If researchers five thousand years from now determine that a large migration of heavily melanistic humans migrated from west Africa to North America in the last millenium, and that although the still resident population of NA clearly showed the survival and continuity of that African ancestry in other respects the skin tone and its supporting genetics had vanished - been replaced by a different skin tone with a different supporting genetic complex among the humans living in North America, so that no population of heavily melanistic humans with African skin tone genetic complexes remained in NA - they would be reasonable (as well as almost certainly correct, eh?) in concluding that evolutionary change under selection pressure had taken place.

 

 

 

I'm not even going to get into how badly you are butchering terminology here, but the fact that all these factors are "unknown to me, and irrelevant to my argument" is disgraceful.

Well, I seem to have no problem with disgracing myself, certainly not in your eyes, so with that established something relevant to an argument or observation that I posted would be a reasonable path to follow in your responses from now on - or a simple avoidance of response, when unable to make sense of things (such as whenever you are tempted to write the word "nonsense" with reference to one of my posts) would be another reasonable possibility open to you.

 

In that way, you can avoid this kind of reponse:

 

2) This: "the genetics are plastic (set up to allow sensitivity to selection pressure, unlike, say, design height or digestive tract physiology)" is nonsense. How are the genetics "set up allow sensitivity to selection pressure"? What is the mechanism of such sensitivity?

which would be better than rereading it later with sudden comprehension.

 

This is the observation and argument, originally just a couple of sentences pointing to the heavy selection pressure avaialble on skin color inland and north through Europe, in its first expanded and excessively detailed form after I took warning that there were some problems out there (note the link to the complexities of the northern Indian caucasian blacks)

Not for such plastic traits (in humans) as skin and eye color. They are among the most easily and quickly altered traits humans possess - so much riding on this plasticity that evolution has established a meta-level, an evolved ability to alter skin tone in response to environment (tanning). There are northern India groups of European origin and lactose tolerance (related to Latvians, iirc) that have become racially US black apparently within recorded history (if early Greek and Roman writings are entered as history).

 

Athough it's complicated: http://www.cam.ac.uk...dian-population

 

Considering that vitamin D deficiency directly affects reproduction at the level of female fecundity, gestation and successful childbirth and early chodhood nutrition, while skin cancer and sunburn's effects kill even the prereproductive young, the evolutionary pressure during the colonization of northern Europe after the most recent glacial era might easily have been dramatic.

[chadn737]

They are not assumptions, but observations. And while of course insufficient for final and incontrovertible "conclusion", those observations are strong evidence - one would not be surprised to discover that the indicated evolution under selection pressure had in fact happened in one, two, three, four, or even all of the events. And this latest one of course fits that pattern.

In fact, looked at from another perspective, the circumstances of the latest discovery might be reasonably taken as supporting evidence for similar hypotheses put forward to explain the many other less easily investigated skin color transformations in humans we encounter around the planet, all within a couple of thousand generations.

They are assumptions. You may have made some observations, but that does not constitute strong evidence. In fact, the lack of population continuity in places like India would actually be evidence against strong selective pressure.

If researchers five thousand years from now determine that a large migration of heavily melanistic humans migrated from west Africa to North America in the last millenium, and that although the still resident population of NA clearly showed the survival and continuity of that African ancestry in other respects the skin tone and its supporting genetics had vanished - been replaced by a different skin tone with a different supporting genetic complex among the humans living in North America, so that no population of heavily melanistic humans with African skin tone genetic complexes remained in NA - they would be reasonable (as well as almost certainly correct, eh?) in concluding that evolutionary change under selection pressure had taken place.

 

 

Seriously, the run-on sentences are nearly impossible to read. Its ok to use periods. I give you my permission and blessing.

 

If they showed population continuity, yes, but thats not the case. I raised that example because there isn't population continuity. The average African American has at least 10% European ancestry and many have much more. When you plot out these populations by their principle component, African Americans span a wide range from "more European" to "more African" in terms of genetic heritage. On a complex genetic trait with many loci, this will have an effect.

 

 

In that way, you can avoid this kind of reponse:

Quote

2) This: "the genetics are plastic (set up to allow sensitivity to selection pressure, unlike, say, design height or digestive tract physiology)" is nonsense. How are the genetics "set up allow sensitivity to selection pressure"? What is the mechanism of such sensitivity?

which would be better than rereading it later with sudden comprehension.

This is the observation and argument, originally just a couple of sentences pointing to the heavy selection pressure avaialble on skin color inland and north through Europe, in its first expanded and excessively detailed form after I took warning that there were some problems out there (note the link to the complexities of the northern Indian caucasian blacks)

Quote

Not for such plastic traits (in humans) as skin and eye color. They are among the most easily and quickly altered traits humans possess - so much riding on this plasticity that evolution has established a meta-level, an evolved ability to alter skin tone in response to environment (tanning). There are northern India groups of European origin and lactose tolerance (related to Latvians, iirc) that have become racially US black apparently within recorded history (if early Greek and Roman writings are entered as history).

 

Athough it's complicated: http://www.cam.ac.uk...dian-population

 

Considering that vitamin D deficiency directly affects reproduction at the level of female fecundity, gestation and successful childbirth and early chodhood nutrition, while skin cancer and sunburn's effects kill even the prereproductive young, the evolutionary pressure during the colonization of northern Europe after the most recent glacial era might easily have been dramatic.

 

 

Nothing you just said explains what you really mean by how the genetics are "set up to allow sensitivity to selection pressure"?

 

Furthermore, the article on skin color in India that you reference proves the point I have made about how you cannot assume selection, particularly strong selection, without knowing all the factors I have described. That article is describing results from this paper: The Light Skin Allele of SLC24A5 in South Asians and Europeans Shares Identity by Descent. One of the main points of that paper is that in South India, where one would presume that selection against light skin is strongest, there is not evidence of selection. In fact, it shows that its explainable by demographic factors.

 

From the discussion:

 

"The lack of a clear latitudinal (North-South) cline in the A allele frequency, which would have been expected under the model of natural selection, could be partly explained by the complexity of the South Asian genetic landscape, influenced by differences in population histories shaped by various micro-level migrations within the subcontinent, strict endogamy and social barriers. For example, Saurashtrians, who migrated from “Saurashtra” region of Gujarat to South India (Madurai) for work, have a relatively high rs1426654-A allele frequency of 0.70. It is believed that those Saurashtrians presently dwelling in Madurai were invited by Nayak kings for their expertise in silk-weaving [36]. Similarly, Toda have higher A allele frequency (0.86) compared to Kurumba (0.20), their geographical neighbors, most likely due to their higher proportion of West Eurasian ancestry which is supported by Y chromosome evidence [37]. Notably, Brahmins, irrespective of their geographic source (North, Central or South India) have higher A allele frequency (Table S5). Conversely, the higher longitudinal correlation could be due to the fact that Tibeto-Burman and Austroasiatic speakers are characterized by very low A allele frequency (Table S6) because of their East Asian ancestry [26], [38]. Therefore, their inclusion in our sampling might have resulted in the inflation of the longitudinal correlation coefficient."

 

My point exactly.

Edited March 20, 2014 by chadn737

[overtone]

Nothing you just said explains what you really mean by how the genetics are "set up to allow sensitivity to selection pressure"?

I'm sorry, but I've posted sufficiently on that very simple matter. Reread or not, at your convenience.

 

 

Furthermore, the article on skin color in India that you reference proves the point I have made about how you cannot assume selection, particularly strong selection, without knowing all the factors I have described. That article is describing results from this paper: The Light Skin Allele of SLC24A5 in South Asians and Europeans Shares Identity by Descent. One of the main points of that paper is that in South India, where one would presume that selection against light skin is strongest, there is not evidence of selection. In fact, it shows that its explainable by demographic factors.

I still have no idea why you think "demographic factors" in themselves imply an absence of strong selection.

 

The rapid complete disappearance or universal establishment of an environmentally sensitive trait from an inmigrating population or its "admixtures", for example, does not imply a lack of selection pressure on the underlying genetics. Somewhat the opposite.

 

To use your example: If we came back to North America in a few hundred years and found everyone white skinned, or everyone black skinned, or everyone red skinned, or everyone in the north white and everyone in the south black, and so forth, "population continuity" would not be required to support the inference of selection for skin color. That's what happened to the caucasoid migrants from Europe to India - their white skin, as impllied by their recently originating population, vanished from the "admixture", replaced by black skin.

Edited March 24, 2014 by overtone

[chadn737]

I'm sorry, but I've posted sufficiently on that very simple matter. Reread or not, at your convenience.

 

 

I still have no idea why you think "demographic factors" in themselves imply an absence of strong selection.

 

The rapid complete disappearance or universal establishment of an environmentally sensitive trait from an inmigrating population or its "admixtures", for example, does not imply a lack of selection pressure on the underlying genetics. Somewhat the opposite.

 

To use your example: If we came back to North America in a few hundred years and found everyone white skinned, or everyone black skinned, or everyone red skinned, or everyone in the north white and everyone in the south black, and so forth, "population continuity" would not be required to support the inference of selection for skin color. That's what happened to the caucasoid migrants from Europe to India - their white skin, as impllied by their recently originating population, vanished from the "admixture", replaced by black skin.

 

That is completely wrong.

 

First off, I never said that evidence of population discontinuity implied no selective pressure. I said that you cannot infer selection just because you observe a change without accounting for all these factors. You in contrast have repeatedly claimed the mere presence of changes as evidence for selection without accounting for these factors or even citing a single selection coefficient. That is completely unsupported nonsense.

 

Secondly, the paper you referenced earlier and which I just referenced clearly states that there was NO selection in south Indian populations. Did you just completely ignore that? They clearly show that skin tone differences were a matter of demographic differences rather than any selection on the skin color itself. It amazes me how you can completely deny that evidence and reassert your claim based on no support. They even said that these results are not at all what one would expect from a model of natural selection.

 

Thirdly, let me cite Pickrell and Reich, two of the foremost human geneticists and evolutionary biologists today:

 

"More generally, it has been observed that allele frequencies at loci under natural

selection in humans tend to track neutral population structure rather than any

obvious geographical variation in selection pressures (Coop et al., 2009; Granka et

al., 2012). One factor contributing to this observation may be that selection

pressures on individual loci are relatively weak due to the quantitative nature of

phenotypes (Hancock et al., 2010b; Hernandez et al., 2011; Pritchard et al., 2010;

Turchin et al., 2012). Another contributing factor may be that population

movements over the last several thousand years have to some extent decoupled the

geographic distributions of selected alleles from the geographic distributions of

selective pressures. From the point of view of an individual allele, the movement of

populations acts as a form of fluctuating selection; as populations move from

environment to environment, the selection coefficient on an allele may change in

both sign and magnitude (assuming a fixed selection coefficient in a given

environment). This means that the environment that dominated the allele frequency

trajectory may not be the environment that the allele is found in today."

This is further reinforced by Joseph Pickrell's own tweets on the paper of European skin color evolution where he agreed with a science blogger on human anthropology the problems of distinguishing selection from migration.

 

"An invocation of selection as an explanation requires evidence population continuity, otherwise changes in allele frequency may involve migration of a new frequency-differentiated new population; for example, the massive change in pigmentation in North America over the last 500 years is not due to selection but to migration of Europeans. The authors cannot reject population continuity on the basis of mtDNA haplogroup frequencies, although autosomal data may be more informative for that purpose."

In other words, you have to be careful not to confuse selection with migration. Hence one of the main reasons for the paper by Pickrell and Reich.

In other words, migration can have the affect of decoupling observed allele frequencies from their actual selective pressures, so that we cannot simply assume that observed traits in one region are caused by selection in that region. We see exactly that in the allele frequencies associated with light skin in South India, have nothing to do with selection on those allele frequencies, but rather are the result of migration patterns and continued mating within that population. For instance the Saurashtrians have high frequencies of the light skin allele rs1426654-A. This has absolutely nothing to do with selection, but the demographic changes in populations that have occurred historically.

For you to assert selection in the absence of knowing actual evidence like selection coefficients is naive.

Edited March 24, 2014 by chadn737

[overtone]

 

 

Thirdly, let me cite Pickrell and Reich, two of the foremost human geneticists and evolutionary biologists today:

You can cite experts all you want, but you have three problems: you seem to misread them as you do others, you are once again arguing your version of my posting (so they often aren't even relevant), and your experts are in some cases simply and obviously wrong.

 

Here's an example of that last:

 

 

An invocation of selection as an explanation requires evidence population continuity, otherwise changes in allele frequency may involve migration of a new frequency-differentiated new population; for example, the massive change in pigmentation in North America over the last 500 years is not due to selection but to migration of Europeans.

1) The pigmentation in NA over the last five hundred years has not seen a replacement - the original pigments are still here, some new ones have been added, there's been admixture, etc. Selection on skin color, if any, is still in progress, in other words. Check back in a couple thousand years, rather than claim no selection.

 

2) Just because a bunch of alleles are not under selection pressure, doesn't mean none are.

 

and worst: 3) The massive changes in skin color frequency and location in NA are in fact due in part to very dramatic and obvious and even famous selection events, some directly associated with skin color and others not. The disease epidemics that preferentially killed off the Reds, the greater ability of some Blacks to survive field work in hot weather malaria and yellow fever zones, the same combination of lactose tolerance and Vitamin D and sunburn vulnerability in White's that settled them inland and north in Europe sending them inland and north in the New World - - WTF is he trying to say there? That when you see a big mixture of traits all in flux, you should not jump in and claim "selection"? OK, we won't do that. Does he notice that in such circumstances - massive population moves, etc - he doesn't know that there isn't selection happening either? Apparently not - he claims, on the basis of no evidence whatsoever, an absence of selection in a situation where one would expect it but be unable to spot it.

 

Notice that where there has been replacement, essentially, of skin color - Caribbean Islands that changed from Red to Black skin tone, say - there was serious selection driving it. Disease resistance, sun damage resistance, When Oliver Cromwell shipped tens of thousands of Irish whites to the sugar cane plantations of the Caribbean, their skin color vanished in less than two generations.

 

Or to repeat:

 

The rapid complete disappearance or universal establishment of an environmentally sensitive trait from an inmigrating population or its "admixtures", for example, does not imply a lack of selection pressure on the underlying genetics. Somewhat the opposite.

[chadn737]

You can cite experts all you want, but you have three problems: you seem to misread them as you do others, you are once again arguing your version of my posting (so they often aren't even relevant), and your experts are in some cases simply and obviously wrong.

 

Here's an example of that last:

 

 

1) The pigmentation in NA over the last five hundred years has not seen a replacement - the original pigments are still here, some new ones have been added, there's been admixture, etc. Selection on skin color, if any, is still in progress, in other words. Check back in a couple thousand years, rather than claim no selection.

 

2) Just because a bunch of alleles are not under selection pressure, doesn't mean none are.

 

and worst: 3) The massive changes in skin color frequency and location in NA are in fact due in part to very dramatic and obvious and even famous selection events, some directly associated with skin color and others not. The disease epidemics that preferentially killed off the Reds, the greater ability of some Blacks to survive field work in hot weather malaria and yellow fever zones, the same combination of lactose tolerance and Vitamin D and sunburn vulnerability in White's that settled them inland and north in Europe sending them inland and north in the New World - - WTF is he trying to say there? That when you see a big mixture of traits all in flux, you should not jump in and claim "selection"? OK, we won't do that. Does he notice that in such circumstances - massive population moves, etc - he doesn't know that there isn't selection happening either? Apparently not - he claims, on the basis of no evidence whatsoever, an absence of selection in a situation where one would expect it but be unable to spot it.

 

Notice that where there has been replacement, essentially, of skin color - Caribbean Islands that changed from Red to Black skin tone, say - there was serious selection driving it. Disease resistance, sun damage resistance, When Oliver Cromwell shipped tens of thousands of Irish whites to the sugar cane plantations of the Caribbean, their skin color vanished in less than two generations.

 

Or to repeat:

 

 

Yes, forget what the experts say. Ignore all the evidence that I have presented. Ignore the very paper that says there has been no selection in South India. You know better based on nothing more than assumptions.

 

Once again, when presented with evidence to the contrary, you turn a blind eye and simply repeat yourself. Argument from repetition.

[overtone]

 

 

Yes, forget what the experts say. Ignore all the evidence that I have presented. Ignore the very paper that says there has been no selection in South India.

I have no opinion on whether the alleles discussed in that paper have been selected against, for, or not at all - whatever the researchers claim is fine with me.

 

The elimination of white skin from the Indian descendents of the inmigrating caucasoids, almost within recorded history, is an observed fact. That trait is under genetic control, is an example of what a lot of people refer to as a part of one's "genetic heritage" - agreed?

 

Skin color in humans has changed several times among the various migrating populations of humans on thsi planet - just within the past few tens of thousands of years.

 

 

 

You know better based on nothing more than assumptions.

I know better than to post this as expertise relevant to this argument::

 

for example, the massive change in pigmentation in North America over the last 500 years is not due to selection but to migration of Europeans.

I mean, do you agree that the massive changes in skin pigmentation in North America in the last 500 years is not due - not at all due - to selection? That selection has demonstrably not been involved and is demonstrably not now happening?

[chadn737]

I have no opinion on whether the alleles discussed in that paper have been selected against, for, or not at all - whatever the researchers claim is fine with me.

 

The elimination of white skin from the Indian descendents of the inmigrating caucasoids, almost within recorded history, is an observed fact. That trait is under genetic control, is an example of what a lot of people refer to as a part of one's "genetic heritage" - agreed?

 

Skin color in humans has changed several times among the various migrating populations of humans on thsi planet - just within the past few tens of thousands of years.

 

Just because skin color changes does not mean it is the result of natural selection. That is an unsupported claim and one you have made many times. Although multiple genes are involved, we know many of them and especially the large effect alleles. That these alleles are not under selection is evidence against selection.

 

I mean, do you agree that the massive changes in skin pigmentation in North America in the last 500 years is not due - not at all due - to selection? That selection has demonstrably not been involved and is demonstrably not now happening?

 

 

Considering that I made this exact argument many posts ago, it should be obvious that I agree. I raised this very example as evidence against your claim of selection.

[overtone]

The elimination of white skin from the Indian descendents of the inmigrating caucasoids, almost within recorded history, is an observed fact. That trait is under genetic control, is an example of what a lot of people refer to as a part of one's "genetic heritage" - agreed?

 

 

Just because skin color changes does not mean it is the result of natural selection. That is an unsupported claim and one you have made many times.

I'm waiting for the alternative explanation for the several specific examples mentioned. And waving your hands and saying "admixture" does not explain, say, (among the half dozen you have yet to explain) the sequential appearance and disappearance of a population-universal genetically controlled trait. What happened to it, is the question you need to address.

 

 

 

Although multiple genes are involved, we know many of them and especially the large effect alleles. That these alleles are not under selection is evidence against selection.

True. Last I checked, 77 known genetic locales have some role in human skin color, and everyone here agrees that any alleles still present and functional in their statistically expected frequencies and combinations have not been under selection.

 

The situation you have yet to address is that of major, visible, obvious, and genetically controlled traits vanishing from entire populations of humans. The statistical odds against this happening by chance, say an admixture that by random recombination replaced a certain set of alleles with a new set throughout the descendents of the population without biasing the frequencies of the other alleles being admixed, are rather long, eh?

 

In the one case of the white skin first appearing and then disappearing from a human population we label "caucasian" otherwise, you have two such vanishing acts to explain: first that of black skin, in the population that migrated to Europe from Africa via the Middle East, and then that of white skin when that same genetically identifiable (and physically - despite admixture their descendents look caucasion, barring skin color) group migrated to India.

 

Now there is an explanation available to you for some of this, and the discovery that several alleles known to be involved in the expression "white skin" remain among the dark skinned caucasians of India in their expected frequencies, mixed in, points to it: that the trait involved is governed almost entirely by a fair number of recessive alleles inherited independently, and one needs the full homozygous set for "white skin". So white skin becomes statistically improbable, rather than selected out.

 

The problem you would have with that argument here is that the rapid development of such white skin in the first place - the event that is supposed to be surprising to us, the OP - becomes essentially impossible as a consequence of "admixture" with only light selection, and much less likely as a feature of an incoming total replacement group (where'd they get it?), and even more shocking in its rapidity - relying as it would on powerful and strictly negative selection on each of the dark skin alleles, without benefit from the invisibly accreting white skin alleles hidden in the population.

 

 

I mean, do you agree that the massive changes in skin pigmentation in North America in the last 500 years is not due - not at all due - to selection? That selection has demonstrably not been involved and is demonstrably not now happening?

 

 

Considering that I made this exact argument many posts ago, it should be obvious that I agree.

Uh, Ohkaaay. Not sure how to deal with that. We do have several major selection events kind of, you know, written down from first hand observation and many years of solid research and so forth - it's hard for me to believe you haven't run across them; and you are the one lecturing us all on "admixtures" etc so you would of course recognize a selecftion event when presented with one - right?

 

However, something tells me we have one of those vocabulary problems going on again - all of a sudden it's "natural selection", a new adjective has appeared, which I have a feeling is not going to include (say) what happened to the Irish Cromwell sent to join the "admixture" in the Caribbean sugar plantations, or the slave children who left home and passed for white in their new communities in the US - is that the explanation for such a goofy assertion? That the diseases and wars taking out 60 - 90% of the Red skin color on the NA continent were not "selection" because they were socially mediated within the species? That the selection of black skin humans for such tasks as - say - digging the Panama Canal, because despite their laziness and unreliability

 

(I'm paraphrasing the head of the American operation, forced to hire Jamaicans despite the fact that a couple of centuries after their original "admixture" of reds, browns, whites, and blacks, the remaining population was all black skinned)

 

they could work in the sun all day and did not get sick and die, doesn't count?

 

Let's say the trends of the past three hundred years continue for another three thousand, and researchers doing DNA analysis of the long dead populations of 24th Millenium Mississippi River Valley hominids (after the Great Quake and depopulation of the central continent) find that the southern ones had black skin with some brown and yellow, and the northern ones had white skin with some brown and yellow, and nobody had red skin. No selection?

Edited March 26, 2014 by overtone

[chadn737]

I'm waiting for the alternative explanation for the several specific examples mentioned. And waving your hands and saying "admixture" does not explain, say, (among the half dozen you have yet to explain) the sequential appearance and disappearance of a population-universal genetically controlled trait. What happened to it, is the question you need to address.

 

You're the one making unsupported assumptions regarding selection overtone. I have made no claim except that you have to account for population discontinuity, demographics, effective population size, etc in order to claim that there has been selection. In the case of South India, the previously linked paper clearly showed that this was not explainable by natural selection, but rather was a factor of demographic shifts and migrations.

 

Lets step back and take a very basic review of population genetics.

 

The skin tone of a population will change if the underlying allele frequency changes. The allele frequency can change due to one or more factors:

 

1) Natural Selection

2) Assortative mating

3) Mutation

4) Genetic Drift

5) Gene Flow (Migration)

6) Meiotic Drive

 

In this circumstance, Mutation is going to be minor to irrelevant as is Meiotic Drive. So the question is, did the allele frequencies change due to Natural Selection or because of the other factors.

 

An important factor here is the effective population size N_e. If N_e is small, then the effects of genetic drift are much more pronounced and allele frequency changes can occur due to random fluctuations alone, no natural selection.

 

Assortative mating, or non-random mating, can also shift the population structure. This can result from cultural/demographic factors. Consider that 8% of men in the former Mongol Empire (0.5% in the world), carry Genghis Khan's Y-Chromosome. This is not because there is selective pressure on the Y-chromosome itself. Genghis Khan had access to any woman he wanted, as did his sons and their sons. Men in that society at the lower end had less access to women. As a result, allele frequencies shifted.

 

Gene flow, should be obvious. Sample the population of South Africa a 1000 years ago, and you would find very low or no alleles associated with white skin. Sample that population again today and you would find a significant number. Was that due to natural selection? No, white Europeans simply migrated there and today make up a significant minority of that region.

 

There are multiple explanations, that is why you never assume natural selection. In population and evolutionary genetics the null hypothesis is one of neutrality. In other words, one tests for selection against the assumption of no selection. A lot of bad research, particularly in genome wide association studies, has been conducted over the years because people failed to account for population structure and other factors that give false association.

 

You, throughout this thread, have failed to account for all these factors. The only source you cited, on skin tone in India even demonstrates the folly of assuming selection without accounting for these factors.

 

 

True. Last I checked, 77 known genetic locales have some role in human skin color, and everyone here agrees that any alleles still present and functional in their statistically expected frequencies and combinations have not been under selection.

The situation you have yet to address is that of major, visible, obvious, and genetically controlled traits vanishing from entire populations of humans. The statistical odds against this happening by chance, say an admixture that by random recombination replaced a certain set of alleles with a new set throughout the descendents of the population without biasing the frequencies of the other alleles being admixed, are rather long, eh?

In the one case of the white skin first appearing and then disappearing from a human population we label "caucasian" otherwise, you have two such vanishing acts to explain: first that of black skin, in the population that migrated to Europe from Africa via the Middle East, and then that of white skin when that same genetically identifiable (and physically - despite admixture their descendents look caucasion, barring skin color) group migrated to India.

Now there is an explanation available to you for some of this, and the discovery that several alleles known to be involved in the expression "white skin" remain among the dark skinned caucasians of India in their expected frequencies, mixed in, points to it: that the trait involved is governed almost entirely by a fair number of recessive alleles inherited independently, and one needs the full homozygous set for "white skin". So white skin becomes statistically improbable, rather than selected out.

The problem you would have with that argument here is that the rapid development of such white skin in the first place - the event that is supposed to be surprising to us, the OP - becomes essentially impossible as a consequence of "admixture" with only light selection, and much less likely as a feature of an incoming total replacement group (where'd they get it?), and even more shocking in its rapidity - relying as it would on powerful and strictly negative selection on each of the dark skin alleles, without benefit from the invisibly accreting white skin alleles hidden in the population.

 

Your argument sounds a lot like a Creationist, pushing back the time frame and asking "where they got it"?

 

Once again you simply assume selection without eliminating other possibilities. Again lets go back to basic population genetics. Remember the effective population size, N_e? If N_e​ is small, genetic drift gets larger. I know you have heard of population bottlenecks. If a region were settled by a very small initial population, taken from a larger population, then you could have very radical shifts in allele frequency from that alone and not due to natural selection. This is exactly what has happened in many islands, Iceland being the best studied.

 

You are going around starting with the assumption of natural selection, yet lack any the necessary evidence to support. I have never claimed that there is not natural selection. I have only argued that your assumptions are groundless until you actually test for selection.

 

 

Uh, Ohkaaay. Not sure how to deal with that. We do have several major selection events kind of, you know, written down from first hand observation and many years of solid research and so forth - it's hard for me to believe you haven't run across them; and you are the one lecturing us all on "admixtures" etc so you would of course recognizea selecftion event when presented with one - right?

However, something tells me we have one of those vocabulary problems going on again - all of a sudden it's "natural selection", a new adjective has appeared, which I have a feeling is not going to include (say) what happened to the Irish Cromwell sent to join the "admixture" in the Caribbean sugar plantations, or the slave children who left home and passed for white in their new communities in the US - is that the explanation for such a goofy assertion? That the diseases and wars taking out 60 - 90% of the Red skin color on the NA continent were not "selection" because they were socially mediated within the species? That the selection of black skin humans for such tasks as - say - digging the Panama Canal, because despite their laziness and unreliability

(I'm paraphrasing the head of the American operation, forced to hire Jamaicans despite the fact that a couple of centuries after their original "admixture" of reds, browns, whites, and blacks, the remaining population was all black skinned)

they could work in the sun all day and did not get sick and die, doesn't count?

Let's say the trends of the past three hundred years continue for another three thousand, and researchers doing DNA analysis of the long dead populations of 24th Millenium Mississippi River Valley hominids (after the Great Quake and depopulation of the central continent) find that the southern ones had black skin with some brown and yellow, and the northern ones had white skin with some brown and yellow, and nobody had red skin. No selection?

 

 

You have no evidence of selection in any of those circumstances. You are simply assuming selection on skin tone without actually having tested for it. In some of these examples, its simply outrageous. For instance, the killing of most of the Native American population was due to lack of resistance to the diseases introduced by Europeans. That had nothing to do with whether or not their skin color was beneficial.

 

The time scale here is also so incredibly short, that its really hard to believe that it was selection. In contrast, migration and admixture...not to mention the demographics imposed by racism...were so overwhelming that it makes for a far more realistic explanation. Actual human geneticists agree, as I showed previously.

[CharonY]

I think the reasoning is based on ignorance. In school, often natural selection is presented as the main force affecting observed allele frequencies. The ignorance part is that it is but one of many possibilities as chadn737 explained. Therefore many may assume that in the absence of knowing the precise events the main explanation must be selection. However, the claim of selection is a highly specific hypothesis that can be tested under certain models. If that fails it is simply inaccurate to assume selection.

In most cases it will be very tricky or impossible to correctly calculate precise contributions of all events on observed allele frequencies. Yet claiming one specific model to be true without any data to support it is at best intellectual dishonesty.

[overtone]

So the question is, did the allele frequencies change due to Natural Selection or because of the other factors.

As with the India example, if you want to draw conclusions about trait disappearance from allele frequency you have to know which alleles are involved and how their distribution affected the trait. Migration and admixture in themselves do not make traits vanish from populations, absent specific properties of the alleles and the mechanism of "admixture".

 

 

 

An important factor here is the effective population size N_e. If N_e is small, then the effects of genetic drift are much more pronounced and allele frequency changes can occur due to random fluctuations alone, no natural selection.

Which is why the example of India needs more than handwaving about "admixture" - the trait disappeared, but a whole lot of relevant alleles remained. There are a couple of different explanations for that (even a selection possibility, say via mating or health effects or a combination of those two factors working on a couple of related alleles not measured by those researchers). If the explanation you do come up with implies recessive alleles throughout, which is what you seem to favor in some vague way, then genetic drift for the creation of the original European population gets much less likely at any given (realistic) founding population size and migration pattern. Too many genes have to drift away simultaneously by chance, and the remaining inbred population has to be viable as well as unusually isolated (you seem to agree with the rest of us that isolation in a human population is unlikely - we wander a lot, and mate everywhere we go, always have)

 

 

 

Assortative mating, or non-random mating, can also shift the population structure.

Assortative mating is a mechanism of natural selection. At least, it is in every other animal, and people like me treat humans as large mammals when discussing evolutionary change.

 

 

 

Once again you simply assume selection without eliminating other possibilities.

No, I don't.

 

You can tell me what I'm assuming when you have managed to follow some dead simple argument I've made without 1) imposing your own idiotic and uncorrectable presumptions and then 2) demanding I provide you with peer reviewed sources for them besides your ass.

 

 

 

Again lets go back to basic population genetics.

. Look, if you want me to go over your population genetics homework and correct your errors, I'll be happy to oblige on another thread. I've done enough of that in my life to have the routine down, and it wouldn't be too much work - you aren't completely at sea here. But absent an argument it's not relevant to anything I've been posting, so don't bring it into replies to my posts on this thread, OK?

 

 

 

The time scale here is also so incredibly short, that its really hard to believe that it was selection. In contrast, migration and admixture...not to mention the demographics imposed by racism...were so overwhelming that it makes for a far more realistic explanation.

The problem with migration and admixture is that they don't in themselves explain the original appearance or disappearance of the traits involved in so short a time - especially in the assumption of an absence of isolation and inbreeding, as you seem to agree describes a more realistic view of human migration patterns.

 

Another issue with your invocation of racism and so forth is that such mate selection and bias in reproductive success normally falls under the umbrella of "natural selection" as biologists who know what they are talking about use the term. Furthermore, we have modern and recorded examples of the operations of such factors, and as one would expect on Darwinian terms they normally mediate or amplify environmental pressures such as disease, vitamin deficiency, and sun exposure skin problems (skin disorders and other signs of immune system malfunction have serious effects on human mate choice, right?). I pointed to a couple above - Jamaica, the Panama Canal, Mississippi. To presume otherwise would require argument, justification.

 

 

 

You have no evidence of selection in any of those circumstances.

We have the diaries and records of the people doing the selecting, in which skin tone is featured prominently. We have the results, which is skin tone dominance after some years of "admixture" in a given region.

 

You are simply assuming selection on skin tone without actually having tested for it.

The difference between "assuming" and "reasoning for" would be a very good thing for you to learn, some time in your life.

 

In some of these examples, its simply outrageous. For instance, the killing of most of the Native American population was due to lack of resistance to the diseases introduced by Europeans. That had nothing to do with whether or not their skin color was beneficial.

Are we agreed, then, that the near and possibly ongoing disappearance of Red skin tone from huge areas of the North American continent was a natural selection event, and not as you and your expert so ridiculously asserted attributable to "migration" without selection?

 

The ignorance part is that it is but one of many possibilities as chadn737 explained.

Please don't do that. Chad giving up on the "explanations" and getting a handle on the argument is only going to be delayed by such encouragement.

 

 

Yet claiming one specific model to be true without any data to support it is at best intellectual dishonesty.

Good thing nobody but maybe Chad, if he ever does join the discussion, is in any danger of doing that, then. Your point?

 

However, the claim of selection is a highly specific hypothesis that can be tested under certain models. If that fails it is simply inaccurate to assume selection.

One doesn't "assume" a specific model of selection to the exclusion of everything else, in any realistic situation. So?

Edited March 28, 2014 by overtone

[hypervalent_iodine]

!

Moderator Note

Alright, overtone, that's enough. Your superior attitude and abrasiveness with other members is unacceptable. Furthermore, we will not allow you to continue to make and persist with claims that are unsupported and / or go against accepted science.

 

Just because other members disagree with you (and can support their positions with rational argument as well as evidence) does not mean that they lack skills in reading comprehension. I suggest that on your return to SFN, you take into consideration the fact that your opinions on matters of science are not correct just because you think they are or because you are extremely persistent in your claims, especially when there is actual evidence that lies counter to your position. Please be aware that should you continue your present behavior following your suspension, we will ban you permanently from posting on SFN.

[CharonY]

It is incredible (or maybe expected) how arrogance is so often paired with ignorance. Just because it is biology it does not mean that one can flaunt one's half-knowledge and get away with it (well to be fair, according to anecdotes Feynman did it for some time as a student, but then he was brilliant). In physics people would at this point rightfully point out: "show me the math".

 

The same should be done here. However a single sentence:

 

One doesn't "assume" a specific model of selection to the exclusion of everything else, in any realistic situation. So?

 

Betrays almost the whole lack of knowledge on the subject matter. It appears that overtone assumes that in order to investigate population dynamics or evolutionary events researchers would dream up a scenario with all possible events (stochastic or not) and other confounding factors and then magically come up with a model that explains everything. That would be a futile exercise at best.

Instead researchers either have a specific hypothesis (e.g. do we see selective forces at work for that trait in that population, e.g. by estimating selection coefficients for certain traits) depending a bit on the type of data you can realistically collect. Then you run your statistic analyses (that, again have to be specific for your question and being suitable for your data set), then you draw conclusions.

The models that you try to fit have to fulfill certain assumptions to be valid, and one thing that Chad pointed out is that the calculated selection coefficient would only be valid if population continuity was assured. The reason (to re-iterate a point) is because the calculations are not correct if migration and admixture (or other factors like random drift) actually occurred, but were not taken into account. Therefore, these studies should include approaches to test whether these may skew the results (if at all possible).

 

The way Overtone argues is what one could consider a type of narrative evidence. Skin color correlates with vitamin D metabolism hence in all cases we got strong selection going on there. This is fine to a point, but then you have to go through the same process as roughly outlined and and actually demonstrate it. Obviously, we do not expect people to actually go out and collect data. However, if you put forward such a hypothesis, it is generally expected to provide some literature to support your assertions.

 

Just to make it clear, biology is science and speculations will only get you so far. One may impress laypersons with that but scientists will tell you :"Data or GTFO".

Anecdotes or inductive reasoning are not acceptable substitutes and this goes for all sciences.

 

Edit: crossposted with modnote.

Edited March 28, 2014 by CharonY

[chadn737]

As with the India example, if you want to draw conclusions about trait disappearance from allele frequency you have to know which alleles are involved and how their distribution affected the trait. Migration and admixture in themselves do not make traits vanish from populations, absent specific properties of the alleles and the mechanism of "admixture".

 

1) As this study accounted for those factors, they can actually make that conclusion.

 

2) Basic population genetics....migration shifts allele frequency.

 

Which is why the example of India needs more than handwaving about "admixture" - the trait disappeared, but a whole lot of relevant alleles remained. There are a couple of different explanations for that (even a selection possibility, say via mating or health effects or a combination of those two factors working on a couple of related alleles not measured by those researchers). If the explanation you do come up with implies recessive alleles throughout, which is what you seem to favor in some vague way, then genetic drift for the creation of the original European population gets much less likely at any given (realistic) founding population size and migration pattern. Too many genes have to drift away simultaneously by chance, and the remaining inbred population has to be viable as well as unusually isolated (you seem to agree with the rest of us that isolation in a human population is unlikely - we wander a lot, and mate everywhere we go, always have)

 

 

1) To be replaced with handwaving about selection? Ha! The study actually tests these factors, no handwaving involved.

 

2) The large effect alleles involved in human skin color are known. There may be small effect alleles, but that is not going to explain large shifts. If selection were a cause, a far more likely explanation is that these alleles were eliminated due to linkage to another allele not involved in skin color. For instance, the decimation of Native American populations due to small pox had nothing to do with selection on skin color, yet it had the effect of greatly reducing the number of Native Americans and any associated alleles.

 

3) I have no idea why you seem to think I favor recessive alleles, I have said nothing on that matter and it is senseless.

 

4) You seem to imply that genetic drift only occurs in small or isolated populations. That is not the case and I have made no claims on that matter.

 

5) Isolation has been a factor of human evolution in many cases. For instance, Jewish populations had relatively little outbreeding in comparison to other populations for much of their history. That is why there is such strong population structure in Jewish people at the genetic level. Even today, there are numerous examples of assortative mating leading to reproductive isolation...Amish, Hutterites, etc.

 

6) The scenarios you proposed are confused and make little sense.

 

Look, if you want me to go over your population genetics homework and correct your errors, I'll be happy to oblige on another thread. I've done enough of that in my life to have the routine down, and it wouldn't be too much work - you aren't completely at sea here. But absent an argument it's not relevant to anything I've been posting, so don't bring it into replies to my posts on this thread, OK?

 

 

About that.....see below:

 

Assortative mating is a mechanism of natural selection. At least, it is in every other animal, and people like me treat humans as large mammals when discussing evolutionary change.

 

 

That is wrong. Assortative mating can be part of natural selection. For instance sexual selection leads to assortative mating. It is wrong, however, to say that all assortative mating is natural selection. Assortative mating can result due to temporal and/or spatial separation. For instance, if humans migrate and settle a new region, interaction with the original population will be reduced or cutoff. As a result, a nonrandom mating will occur. Individuals in one region will tend to mate with people from that region.

 

The problem with migration and admixture is that they don't in themselves explain the original appearance or disappearance of the traits involved in so short a time - especially in the assumption of an absence of isolation and inbreeding, as you seem to agree describes a more realistic view of human migration patterns.

 

 

Large scale migration of Europeans to North America don't explain the appearance of white skin in North America? Seriously?

 

Elimination of Native Americans by Europeans doesn't explain their reduced presence? Seriously?

 

You don't need isolation and inbreeding in such cases.

 

Another issue with your invocation of racism and so forth is that such mate selection and bias in reproductive success normally falls under the umbrella of "natural selection" as biologists who know what they are talking about use the term. Furthermore, we have modern and recorded examples of the operations of such factors, and as one would expect on Darwinian terms they normally mediate or amplify environmental pressures such as disease, vitamin deficiency, and sun exposure skin problems (skin disorders and other signs of immune system malfunction have serious effects on human mate choice, right?). I pointed to a couple above - Jamaica, the Panama Canal, Mississippi. To presume otherwise would require argument, justification.

 

 

If assortative mating based on racism were simply an amplification of environmental selection, then one would not expect segregation in these regions.

[Arete]

Ultimately, this comes down to something that I learned very early on in fundamental population genetics - tests of selection can return false positive results if the assumptions of neutrality are violated. There is a large suite of demographic parameters - migration, deviation from HWE, fluctuations in population size, trait auto-correlation, recombination, linkage disequilibrium, founder effects, etc etc which can generate a seemingly significant signal of selection.

 

What this means is that, even if you pulled a bunch of e.g. melanocortin gene sequences from Genbank and tested for selection using Tajima's D, MacDonald-Kreitman, etc, you could wind up with a statistically significant result, even if selection is not present. Meeting the assumptions of such a test is fundamental - which is why a summary statistic like Tajima's D would generally be reported with a suite of other parameter tests (e.g. tests of recombination, LD, theta, Ne estimates, migration estimates, etc) to rule out the influence of demography.

 

In the current case, we don't even have any statistical results, simply speculation about the role of selection on a trait. As such, it is highly relevant to suggest the role of demographic parameters, as they absolutely have not been ruled out and in fact are likely to be influential in generating any speculated signal of selection that is present - if indeed it even is.

 

Again, it's pretty basic pop gen that the role of selection can only be interpreted in the context of demography, and that even if you have a statistically significant result in a test of selection, additional analyses are required to correctly interpret the result.

[chadn737]

Ultimately, this comes down to something that I learned very early on in fundamental population genetics - tests of selection can return false positive results if the assumptions of neutrality are violated. There is a large suite of demographic parameters - migration, deviation from HWE, fluctuations in population size, trait auto-correlation, recombination, linkage disequilibrium, founder effects, etc etc which can generate a seemingly significant signal of selection.

 

What this means is that, even if you pulled a bunch of e.g. melanocortin gene sequences from Genbank and tested for selection using Tajima's D, MacDonald-Kreitman, etc, you could wind up with a statistically significant result, even if selection is not present. Meeting the assumptions of such a test is fundamental - which is why a summary statistic like Tajima's D would generally be reported with a suite of other parameter tests (e.g. tests of recombination, LD, theta, Ne estimates, migration estimates, etc) to rule out the influence of demography.

 

In the current case, we don't even have any statistical results, simply speculation about the role of selection on a trait. As such, it is highly relevant to suggest the role of demographic parameters, as they absolutely have not been ruled out and in fact are likely to be influential in generating any speculated signal of selection that is present - if indeed it even is.

 

Again, it's pretty basic pop gen that the role of selection can only be interpreted in the context of demography, and that even if you have a statistically significant result in a test of selection, additional analyses are required to correctly interpret the result.

 

Exactly. Thank you for explaining that because I fear that my explanations got bogged down in some nastiness.