No genetic bottleneck proves no global flood

Biologists do not expect new phenotypes to pop into existence after a bottleneck and the two extreme examples of the rabbits and the deer demonstrate that. Similarly cheetahs, seals and bison are still cheetahs, seals and bison.This is a factual thing, Faith; if you have examples of new phenotype following a bottleneck, please produce them. Yes, reproductive rate is identified by both papers as likely being the reason that both the deer and the rabbits recovered gentic diversity. Note that Faith, recovered genetic diversity - not developed new phenotypes. Nope. Not unless there is selection pressure or long term drift which happen in populations that have not undergone a bottleneck just as much as those that have. Well of course, that's a mathematical certainty! In the case of the rabbit, genetic diversity recovered to the extent that there was no difference between the host European population and the new Australian population. In the deer, genetic diversity recovered to almost the same level as the Oklahoma population within 70 years - although still different enough to show a bottleneck. (H=.692 vs H=.742). btw, in your earlier post you described the rabbit's H of .5 as 'enormous'.

A little overstated imho. Small phenotypic change - fur color, length, curliness, some increase or decrease in size, for example - can occur and would not be unexpected. Significant new phenotypes, such as new subspecies, which is what Faith is thinking, are not expected, they could occur but don't have to occur without selection pressure for such changes.

I'm thinking small phenotypic changes, and as you agree, those should be expected from new gene frequencies -- or, as you say, "would not be unexpected" though I think they should actually be expected.The idea that I'm expecting some large "significant" change, no, I'm just expecting typical microevolution of external traits. But I'm thinking of subspecies as simply a new population that has been worked through by the new gene frequencies over enough generations to get a characteristic appearance to the group. I don't consider that a "large" change. Nothing new is added, it's just the gene frequencies doing their thing.Apparently there are some species where this doesn't happen, such as rabbits, due to their rapid reproductive cycle and their habit of forming small groups instead of mixing their genes by breeding through the main population as, say, herd animals would do.The numbers Tangle gave for percentage of heterozygosity I'm going to have to spend some time on, because they are very large compared to my understanding that the human percentage is only around 7%. Edited by Faith, : No reason given.Edited by Faith, : No reason given.Edited by Faith, : No reason given.Edited by Faith, : No reason given.

Yes, .5 IS enormous since I've been thinking the typical percentage must be in the neighborhood of the human percentage of 7%.Selection pressure is not needed when you have such a significant bottleneck which would produce dramatically new gene frequencies. To which genetic drift would also contribute, but apparently not in the case of the rabbits according to that article as their habits mitigate that effect.

Within a stable ecology selection would be towards the population average, both by survival and by reproduction. This is why stasis occurs. Thus any changes would necessarily be relegated to fairly minor aspects that would not decrease survival or reproduction, such as the types of variations I mentioned. A new subpopulation is a larger change in phenotype than I would expect, as it means that some selection is occurring to isolate variations into a breeding group.That you don't consider it "large" is inconsequential to me, seeing as it is just your opinion, based on a fantasy that you have concocted which is not only unsupported, but the actual evidence points the other way: that such changes are due to mutations not pre-existing hidden never before seen alleles.Let's take another example: the Galapagos\Darwin finches as studied by the Grants:Evolution: Library: Finch Beak Data Sheet

quote:

---

The Grants wanted to find out whether they could see the force of natural selection at work, judging by which birds survived the changing environment. For the finches, body size and the size and shape of their beaks are traits that vary in adapting to environmental niches or changes in those niches. Body and beak variation occurs randomly. The birds with the best-suited bodies and beaks for the particular environment survive and pass along the successful adaptation from one generation to another through natural selection.

​

Natural selection at its most powerful winnowed certain finches harshly during a severe drought in 1977. That year, the vegetation withered. Seeds of all kinds were scarce. The small, soft ones were quickly exhausted by the birds, leaving mainly large, tough seeds that the finches normally ignore. Under these drastically changing conditions, the struggle to survive favored the larger birds with deep, strong beaks for opening the hard seeds.

​

Smaller finches with less-powerful beaks perished.

​

So the birds that were the winners in the game of natural selection lived to reproduce. The big-beaked finches just happened to be the ones favored by the particular set of conditions Nature imposed that year.

​

Now the next step: evolution. The Grants found that the offspring of the birds that survived the 1977 drought tended to be larger, with bigger beaks. So the adaptation to a changed environment led to a larger-beaked finch population in the following generation.

​

Adaptation can go either way, of course. As the Grants later found, unusually rainy weather in 1984-85 resulted in more small, soft seeds on the menu and fewer of the large, tough ones. Sure enough, the birds best adapted to eat those seeds because of their smaller beaks were the ones that survived and produced the most offspring.

​

Evolution had cycled back the other direction.

---

Note that only beak size changed, and this is a fairly minor phenotypical change.These are larger changes to the phenotypes than we should expect for the Australian rabbits, because (a) the populations were smaller and (b) the ecology changed to the point that small beaked finches could not survive, and it was only when the ecology changed back that small beaks reappeared.This is also similar to the Peppered Moths -- shifting populations in response to ecological stress. And a lack of selection pressure.There is still gene flow between family groups as mates are usually taken from outside families, and this leads to gene flow throughout the breeding population and prevents in-breeding. I'll let you and Tangle sort this out.Edited by RAZD, : ..

Selection pressure may be reduced but it would still likely be prevalent. The major pressure would be for survival, as reproductive selection would be reduced. Which normally reduces available genotypes. particularly the rarer alleles. The gene frequencies would match the original population if the selection pressure is reduced, especially for sexual selection: alleles that do not exist cannot be selected, and normally would most likely not include rare alleles from before the bottleneck -- hence the expected reduction.My feeling is that if the rabbits came from a domesticated stock (highly likely imho) that then selection for domestic use has already reduced variability before going to Australia, and thus would not be much affected compared to the original domestic stock.

I wouldn't expect anything new from a sub-population would you?Take two mice out of a population and I'd expect them to produce mice with the characteristics of some of the mice within the population. These differences may become more fixed with time if they're not deleterious - like we expect more red headed Scots than English redheads (caution, maybe apocryphal!).But, as you say, these irrelevant differences in average phenotypes is not what Faith is looking for, she thinks that species level changes come from isolation and bottlenecks almost instantly and biology knows that it doesn't.Plus, the examples we have of even severe bottlenecks - rabbits, deer, cheetah, bison, seal haven't produced materially different phenotypes. I wonder if it's possible to tell the difference between an Austarlian rabbit and a French rabbit - even with a DNA sample?Edited by Tangle, : No reason given.

You aren't thinking of the effect of changed gene frequencies, nor, apparently, of the fact that a sub-population would certainly HAVE changed gene frequencies. Recognizable changes from the pattern of the population aren't going to show up in the first generation, possibly not until after a few generations, and then it might only be that one phenotype that showed up from time to time in the larger population becomes characteristic of the new population, but along with other traits as well so that a new "look" is created. Even raccoon subpopulations develop slightly different recognizable traits from others although they are still certainly raccoons. In a sense there is nothing more common than this phenomenon I'm talking about. Yes, that's the basic idea. But it shouldn't just be a change in a single trait. Hair type, eye color, skin color etc., even bodily structure, may also become characteristic as the population inbreeds. But these "irrelevant" differences are EXACTLY what I'm looking for. What YOU think "species level changes" would be is NOT what I think they would be, which is one reason I keep objecting to the usual definition of "species" and "speciation." I think what are called by those names are nothing but such small changes brought about by the usual microevolutionary processes that have proceeded to the point where there is enough of a genetic mismatch to prevent continued breeding with the former population. I don't think that takes much, just a series of phenotypic changes of what you call the "irrelevant" sort, from the new gene frequencies, that accumulate in the population over a number of generations until there is that genetic mismatch with former populations. I'm suggesting this can happen when the new population's genetic diversity is appreciably reduced from that of the former populations. "Materially different" means what? I'm not looking for big changes, remember, just a new "look," like the difference between the black wildebeests and the blue. The different herds are recognizably different herds with superficlally different characteristics, but they are still clearly wildebeests.And again, from this discussion it seems that rabbits might not develop a subpopulation characteristic because of their habit of forming small colonies rather than inbreeding within the larger population.

Er, yes. This is what biologists call the theory of evolution: this is how new species are formed. It doesn't happen because the critters genetic diversity is reduced - if it did we would see it every time we buy a couple of hamsters.It happens because of mutation and/or by genetic drift. And for complex beasts, it happens over a long period of time.What's more, the evidence we've seen so far shows that for some, even catastrophic losses of diversity can be recovered from very quickly - all that seems to be necessary is that the animals can breed freely.Edited by Tangle, : No reason given.

What is "new" ? I would expect some variation/s outside an "average" phenotype, but minor (does eye color affect reproduction? survival?) but most mutations would be neutral.And as I've noted before, in a stable ecology I expect selection to favor the "average" phenotype as preferred sexual selection. Curiously, I was doing some reading last night on heterozygosity, and one article said that the predicted\expected heterozygosity from two individuals was 0.75. It seems that the way this is measured does not really quantify what we think it does from a first look, because of the way it deals with population size. And it doesn't seem to count the degree of genetic variation in the parent population.With two mice you would have somewhere between 2 and 4 possible alleles at each loci, depending on whether the two copies of DNA strings in each individual are the same or different (one maternal the other paternal). If we take an off the cuff "average" value for all alleles at 3, then 3 out of a potential 4 would be 0.75. Exactly, as you can see from her reply. Part of this is her belief that there are hidden alleles tucked away that are somehow suppressed in normal population reproduction but become freed to act in small populations. It's her alternative explanation to mutations creating new alleles. And the starling population in the US compared to the UK -- introduced in NY and spread to the whole US is

But nobody denies that evolution occurs, and I'm glad that you agree that I've described it correctly.Reduced genetic diversity does NOT show up in a random pair of hamsters. In fact they seem to have pretty good genetic diversity, enough to produce quite a variety of fur colors and patterns at least, They'd have to be like the cheetah with fixed loci, that is, homozygosity at all the crucial gene loci, to exhibit reduced genetic diversity.Genetic drift operates the same as selection, or the migration of individuals like the rabbits, which results in a new subpopulation, only genetic drift changes the population from within rather than by migration away from it or by the active selection of individuals. All these processes lose alleles over time, which is how reduced genetic diversity comes about.As for the long time required, even for complex beasts it shouldn't take more than a couple or three hundred years to completely blend the gene pool of the isolated population. That's about how long it has taken to establish certain cattle breeds to the point of "breeding true" throughout the whole population, which requires fixed loci for the peculiar characteristics of the breed. Yes, they are specifically bred for their characteristics but over two hundred years the methods would have changed and at first they were probably left to themselves, and with herds that go back hundreds of years even more likely left to themselves. Which is how it happens in the wild, herd animals that continue to herd together developing a characteristic "look."Although I've been ready to grant that rabbits may have habits that at least protect their genetic diversity, there is no way genetic diversity can be recovered under the circumstances of a bottleneck, and that study of deer has raised the question in my mind what on earth they are measuring when they give such high estimates of heterozygosity. I asked way back there how heterozygosity at microsatellite loci has anything to do with the heterozygosity that is needed at specific genes so that they can continue to vary, and I guess nobody knew the answer, but when deer are now discussed as recovering their genetic diversity I'm extremely skeptical about the method employed. I'd like to know: How do human beings stack up for percentage of heterozygosity as measured at microsatellite loci, and cheetahs too. As I've understood it humans have about 7% and cheetahs with their high percentage of fixed loci must have much less.Edited by Faith, : No reason given.Edited by Faith, : No reason given.Edited by Faith, : No reason given.

Yes, the new look of the new population isn't new in the sense that nothing like it has ever appeared before, but it's a new combination of alleles that may affect a number of traits so that where the original population was brown and long haired and green eyes with short noses and ears and tails, the new population after a number of generations of blending may be mottled short haired with blue eyes and long noses, ears and tails and a larger body and so on. It's going to be recognizably different. A new race, "species" or subspecies.I'm thinking standard Mendelian combinations of course, not mutations. Since most mutations ARE neutral there's no reason to expect them to produce new combinations, but there are plenty of built in alleles that can do just that.Some traits are governed by more than one gene so that simple dominant/recessive isn't the whole picture, but a more complicated interaction. Perhaps you are right about that, but there have to be traits to be selected, and that's what normal sexual recombination of existing alleles does, and after the selection individuals with the favorable genetic combinations will pair up more frequently and so on.. This could use some more information and discussion and I've been doing some searches. I'm not sure what you are saying here, but the heterozygosity that COUNTS is at the EVOLVING LOCI, and I can't tell from any of the discussion about these things so far if that's even taken into account. The cheetah's heterozygosity must be near zero because of all their fixed (homozygous) genes for instance. Again the question is WHERE is this seen, at what gene loci, how many gene loci and so on. Four alleles per gene for two individuals is the highest you can get. Three is good, yes, but then you have to note how many genes can be described as being that high. If five genes govern a trait and the same two individuals also have four between them that's enormous heterozygosity for that trait. The question remains WHERE is this heterozygosity happening in the genome and how many gene comparisons are involved and so on? The original articles posted by Tangle on rabbits and deer said the heterozygosity is measured at "microsatellite loci" but there has been no explanation of why this is the focus and what it actually shows. He has obviously misread something but he doesn't quote me and all I can say is that he's wrong about what I've been saying. I do think these changes come from isolation alone, but not "instantly" although the new gene frequencies are of course there from the founding of the new population, but they have to be subjected to sexual recombination through some number of generations to produce a new look to the population. And I've only used bottleneck as an example to demonstrate the extreme, otherwise the trend is the same but slower. I never said it happens "almost instantly" because I know that the new population has to inbreed for a few generations at least to bring out new traits, and if the population started out with a relatively large number of individuals, as many as a hundred, say, then I know it's going to take longer for that to happen. I have NO idea how you are getting such an idea out of my reply. I don't think I'm describing anything hidden or strange at all, just that new gene frequencies is likely to mean that recessives that weren't expressed in the former population may occur more frequently and therefore pair up more frequently in the new, but the same could be said of dominants if recessives characterized some traits in the other population and so on, all this being normal results of normal sexual recombination. And where a number of genes govern one trait you COULD get something that had only showed up very occasionally in the first population, but new combinations of many different traits should eventually produce something that looks appreciably different from the former population. By standard methods of genetic combination, nothing unusual.

Curiously, I am reminded of Monty Python's Flying Circus "and now for something completely new" ... In "standard Mendelian combinations" each of these traits would have appeared in the general population, not just rarely but rather commonly for the alleles to be so well distributed in the population that any small contingent would have them.And this also means that there would likely be combinations of blue eyes and mottled short haired, etc.Furthermore, in "standard Mendelian combinations" there is no mathematical reason for these to suddenly all become sorted into certain individuals that would then become reproductively isolated within the rest of the population sufficiently to form such a sub-population. Nor would there be an evolutionary cause for this - there would be no positive selection for such combinations in an ecology with static selection pressure, and sexual selection would much more likely act to reduce the numbers of such random occurrences of such mathematically rare combinations.

The article I was reading was: Population Bottlenecks: Heterozygosity vs. Allelic Diversity

quote:

---

Question: How do we measure loss of genetic variation? Do the different measures yield similar estimates of the amount of variation lost?

​

Variables:

​

H	heterozygosity
N	population size during bottleneck
n	original number of alleles
n'	number of alleles remaining after bottleneck
pj	frequency of the jth allele
A	allelic diversity
t	number of generations

​

Methods: The two measures of genetic variation examined by Allendorf (1986) were heterozygosity (proportion of individuals heterozygous at a locus) and allelic diversity (the actual number of alleles present at a locus). The expected heterozygosity following a bottleneck lasting a single generation is expressed as a proportion of the original heterozygosity:

​

1 - ¹/_(2N)

​

As N (the population size during the bottleneck) increases the second term (1/[2N]) decreases, and the proportion of the original heterozygosity remaining increases. In general, a bottleneck lasting one generation has a fairly small effect on heterozygosity; even if the population is reduced to 2 individuals the expected proportion of heterozygosity is 0.75 (1 - 1/[2*2] = 0.75). ...

---

This is a rather rudimentary estimate of expected heterozygosity (proportion of individuals heterozygous at a locus), basically like an order of magnitude calculation, as it doesn't include the number of alleles involved.Note that this is looking at only one locus, one gene, and the number of alleles for that one locus\gene. The number 2N is the number of DNA strings in the population size N, and the maximum possible number would be obtained when there is a different allele in each DNA strand for that locus\gene.Next they look at how this stacks up for different numbers of alleles in the original population for the various loci, and the math, unfortunately, gets a little more involved:

quote:

---

... Calculating the expected number of alleles present at a locus is more complicated, because it depends on the number and frequencies of alleles present before the bottleneck. For a diploid locus with n alleles, the frequency of the j^th allele is expressed as p_j (for our purposes we will assume that the alleles are equally frequent; if n = 4 then the frequency of each allele will be 0.25). The number of alleles we would expect following a bottleneck (n') is calculated by the following equation:

​

E(n') = n - Σ_(j=1→n){1 - p_j}^2N

​

For each of the n alleles, we subtract the frequency of that allele from 1 and raise the difference to the bottleneck population size (N) multiplied by 2. ...

---

To break it down, n is the original population number of alleles at a loci, p_j is the frequency of the j^th allele at that loci,and (1-p_j)^2N is the probability of that allele being in the number of DNA strands (2N) of the bottleneck population,The probability of each allele, from 1 to n, is calculated and then they are all added up and this sum is subtracted from the original population number of alleles at a loci, n, to derive the expected number of alleles, n', in the bottleneck population.Applying this to a simplistic example where N = 2 again, and using n = 2, with each allele having the same frequency (p_j = 0.5) in the original population, we get:

1. the probability of allele #1 being in the number of DNA strands (2N = 4) of the bottleneck population:
   (1-p_j)^2N = (1-0.5)^2N = 0.06 (rounded)
2. the probability of the other allele is the same (0.06)
3. the sum of these probabilities is 0.12, so
4. the expected number of alleles (n') after the bottleneck is:
   n' = n - 0.12 = 2 - 0.12 = 1.88

With N = 2 and n = 3 we get n' = 2.41
With N = 2 and n = 4 we get n' = 2.73And this is STILL only one loci of the genetic sequence \ DNA strands \ genes.

quote:

---

... These values are summed for all n alleles, and the sum is subtracted from the original number of alleles (n). We can then use this number (n') to calculate the allelic diversity (A) after a bottleneck. A is the ratio between [the remaining number of alleles minus one] and [the original number of alleles minus one]:

​

A = (n'-1)/(n-1)

​

When all alleles are retained, A = 1. If only a single allele remains, A = 0. ...

---

For our first example this would be: A = (1.88-1)/(2-1) = 0.88
For our second example this would be: A = (2.41-1)/(3-1) = 0.70
For our third example this would be: A = (2.73-1)/(4-1) = 0.58So there is significant variation depending on the numbers of alleles. For just one loci.

quote:

---

... The graph below (redrawn from Allendorf, 1986) illustrates the relationship between allelic diversity remaining after a single-generation bottleneck and the original number of alleles for different population sizes. The black dots indicate the expected proportion of heterozygosity remaining for each population size.

​

​

Note that although the proportion of heterozygosity retained for a bottleneck of N = 2 is 0.75, the proportion of allelic diversity retained when five or more equally frequent alleles are present before the bottleneck is less than 0.50.

---

Certainly when the original number of alleles, n, is greater than the number of DNA strings in the bottleneck population, 2N, there has to be some loss in allelic diversity, as it is mathematically impossible to have more alleles than strings.From what I can see the black dots for the expected heterozygosity are just pasted on the curves at their values and don't have any calculated relationship to the original number of alleles.And we can see that the proportion of heterozygosity is different from the expected allelic diversity, and the larger the number of original alleles the greater this difference will be.Further, to get an idea of the overall effect these calculations would need to be repeated for every gene\loci and the results averaged. A lot of calculations.They go on to calculate the effect over several generations (see graphs) and conclude:

quote:

---

Interpretation: As you can see from the above graphs, in most cases (with the exception of two equally frequent alleles originally present) the expected proportions of heterozygosity remaining (E(H)) indicate a higher retention of genetic variation than measured by allelic diversity (A). It is also apparent that the effects of bottlenecks on genetic variation are greatly increased as population size (N) decreases, and that the duration of the bottleneck (t) compounds these effects.

​

Conclusions: Heterozygosity can be used as a measure of a population's capacity to respond to selection immediately after a bottleneck. Allelic diversity, on the other hand, determines a population's ability to respond to long-term selection over many generations, and ultimately the survival of the population (perhaps even of the species). ...

---

So the expected heterozygosity is best used for initial evaluations, and more long term evaluations should use allelic diversity for a more accurate picture.Genetic diversity, which would include evaluation for each gene\loci and integrating them into an overall picture would be even more complex a calculation.If you have g genes do you multiply all the A values and then take the g^th root? Do you average all the A values? Or is there some other metric that is used?I don't know yet.Edited by RAZD, : formula instead of hard to see imageEdited by RAZD, : ..

Wake me up when you do ;-)