Uncommon Descent Serving The Intelligent Design Community

Kimura and the Adriatic Lizards

Share
Facebook
Twitter
LinkedIn
Flipboard
Print
Email

Over at Panda’s Thumb, they are taking issue with the values for selection probabilities of neutral and advantageous mutations that Sal has taken from Kimura and Ohta’s “Theoretical Aspects of Population Genetics”. Since there was a link that provided a ‘look-see’ inside the book, I did so. Well, what I found was very fascinating.

Kimura and Ohta give a very brief overview of the entire field of population genetics up to the time of their writing (1971), distinctly admiring the pioneering work of R.A. Fisher, but not following it because it uses a more sohpisticated “branching process”, and because his model assumes an “infinite” population size. So they write the following:
“. . . [Haldane’s] results allow us to make statements as ‘it takes about 1,000 generations until the gene frequency changes from 0.7% to 99.3% with selective advantage s= 0.01’. . . .

“More than 30 years after publication of Haldane’s paper (1927b), we have finally begun to understand more about the fate of individual mutant genes in terms of the powerful diffusion methiods based on the Kolmogorov forward and backward equation (cf. Kimura 1964). In particular, the average number of generations until extinction, and also the time until fixation of an individiual muatant gen in a finite population have been workd out (Kimura and Ohta 1969a,b).”

Kimura and Ohta go on to give the kinds of equation that they have developed over the prior 15 years when dealing with fixation. All of this brought to mind a thread I posted just a few weeks back about the amazing phenotypic changes that had taken place in lizards transplanted from one Adriatic island to another just 36 years ago.

Kimura and Ohta, in this highly regarded work on population genetics, first give Haldane’s number of 1,000 generations for the time of a new beneficial mutant gene to become fixed, then talk briefly of the advances made since Haldane’s estimates (we hear this in the quote above), and then, finally, give their own equation.

Their equation is: K=4Nes1v= the rate of gene substitution, where Ne=effective population, s1=selective factor for the beneficial allele, and v=nu/2N=mutation rate/gamete/generation. So, to demonstrate the improvement (i.e., the speedier rate for fixation) of their formulation over Haldane’s estimate, they work out an example. They use Ne- 10^4, s1=0.01, and v is worked out using the rate of deleterious mutations in Drosophila.

Hence, they write: “If we assume that advantageous genes occur by mutation only 1/1000 as oftne as lethal genes, i.e., v=1.5 x 10^-5, and still assuming Nsube=10^4 and ssub1= 10^-2, then we have K= 400 x 1.5 x 10^-5 = 6 x 10^-3 or about one substitution every 170 generations.” They conclude saying: “It is evident from these considerations that the formula for the probability of gene fixation has important applications in evolutionary theory.”

Well, the transplanted population size of the Adriatic lizards was 12. Obviously it grew over time. Let’s see how Kimura’s numbers work out. Let’s assume that Navg is 250. Ne is 0.8 x N = 0.8 x 250= 200. Let’s just assume that ONE mutant gene was fixed (although we know from the paper that many more fixed genes were involved), and let’s assume that it took the entire 36 years to become fixed. Then, using Kimura’s formula, let’s calculate what s1, the selection factor, is.

1/36=K x 36= ONE fixed mutant gene= 4Nes1v. Using v= 1.5 x 10-5, and Ne=200, then solving for s1 we have: s1=1/36 x 200 x 1.5 x 10^-5 =10^5/36 x 300= 100,000/10,800=9.3. But, of course, s cannot be greater than 1. In fact, it can’t even be 1 since that would mean the entire population died out. Thus, the simple math demonstrates that, using the Modern Synthesis, what was documented to have occurred with the transplanted lizard population cannot even begin to be explained using the Darwinian account.

Of course, this is not how Darwinists see things. They don’t look at the evidence, then look at the numbers, and then conclude that obviously something other than Darwinian mechanisms are at play. They just simply say, “Darwinism is a fact. It’s even more than a fact than gravity itself. Therefore, Darwinism is what caused this transformation. Now I’ll just simply look for some kind of facile explanation.” If you don’t believe me, then just watch some of the posts that we all know are coming!!!

It’s the inadequacy of Darwinism to explain matters such as these that led Kimura to his “Neutral Theory”. It also led Sir Fred Hoyle to completely dismiss Darwinian mechanisms. It was my exploration of the mathematics involved that convinced me that Darwinism has to go.

Comments
jerry:
I fail to see what such analysis as this gets ID. So we indicate that certain formula are not perfect. So what. That is what science is about. None of this undermines the Darwinian paradigm or helps ID.
Maybe you should open your eyes. Firstly, it points out the failure of Darwinism. Despite anything that MacNeil wants to say, the Modern Synthesis is THE argument for Darwinism. As I pointed out to Dr. MacNeil when we discussed this some time back, once the MS is destroyed, so too is the mathematical foundation for Darwinism. What is worse, any explanation for the the formation of information is also destroyed. If there is no Darwinian mechanism for explaining the presence of information in the genome, then how is it to be explained? Does this not open the door to ID, jerry?
Micro-evolution is with us and generates most of life’s changes. Let’s embrace the processes that account for it which include natural selection, genetic drift, gene flow and a whole host of other processes that affect off-spring. Let’s get ID out of a rut as sniping at meaningless inconsistencies.
jerry, maybe you would like to explain just exactly how "random drift", "natural selection" and "gene flow" can explain this profound morphological change in a population that began just 36 years ago with an initial population of 10 lizards. Would you like to do that? Remember that cecal valves are found in only a tiny portion of lizards, and had never been seen before in these particular lizards.PaV
May 9, 2008
May
05
May
9
09
2008
07:48 AM
7
07
48
AM
PDT
By trying to point out that the modern synthesis fails, you are slaying a straw man. The current synthesis or whatever you want to call it contains many more things than simple gene selection, genetic drift and gene flow. Evolutionary biologists are looking at numerous other things that affect both genetic and morphological changes. None of which make selection suspect at all. Selection may not explain the immediate changes that take place but it will explain a lot of the long range changes that set in. I am far from an expert in any of this but basic common sense will tell you that there are wide variances within a populations and some of these will affect success in survival. Many of these variance are not visible as morphological differences but still affect survival. They could be enzyme availability or something similar. I fail to see what such analysis as this gets ID. So we indicate that certain formula are not perfect. So what. That is what science is about. None of this undermines the Darwinian paradigm or helps ID. Micro-evolution is with us and generates most of life's changes. Let's embrace the processes that account for it which include natural selection, genetic drift, gene flow and a whole host of other processes that affect off-spring. Let's get ID out of a rut as sniping at meaningless inconsistencies.jerry
May 9, 2008
May
05
May
9
09
2008
07:31 AM
7
07
31
AM
PDT
Sal, don't you think this post is really about Haldane's Dilemna? We discussed this before when it came to Biston betularia, the peppered moth, and I remain convinced that Haldane wrote the paper we discussed in response to the Kettlewell experiment since he had previously estimated that it would take a 1,000 generations to fix a mutant and the Kettlewell experiment was forcing him to rethink things since the phenotypic change had taken place in so few generations. But we saw the change of only one simple trait in the case of the peppered moths, whereas here we have a slew of morphological and behaviorial changes in an incredibley small period of time, with a very small initial--and final--population. This, indeed, is a dilemna for the Modern Synthesis. I'm suspect that's why Allen MacNeil says that the Modern Synthesis is dead. But, of course, if the Modern Synthesis is the foundation of Darwinian theory, then if it fails, why, then, should we take Darwinian theory seriously any longer?PaV
May 9, 2008
May
05
May
9
09
2008
05:59 AM
5
05
59
AM
PDT
Look at Larry Moron's post here: Visible Mutations and Evolutions
Richard Lewontin uses the example of the Indian and African rhinoceros to focus the debate. The African rhinoceros has two horns while the Indian rhinoceros has only one. The question is whether this difference is due to natural selection—is two horns better than one in Africa? Or, is it just an accident of evolution that one species has two horns while the other has only one?
Answer: probably not selection.
I don't understand why the adaptationist camp is so reluctant to admit that some visible characters can be fixed by random genetic drift. The idea that every feature of an organism has to be an adaptation seems so out of touch with our modern understanding of evolution that I'm really puzzled by the vehemence with which adaptationists defend their orthodoxy. It seems as though admitting that visible phenotypes might be non-adaptive is a major threat to their worldview. Larry Moron
scordova
May 8, 2008
May
05
May
8
08
2008
07:58 PM
7
07
58
PM
PDT
dreamwalker007: It's on page 12. On the same page Kimura makes a calculation of K. The value he arrives at is 6 x 10^-3---that is, .006.
There’s a number of alarm bells going off for me on this post. Including the fact that most of the paper around where the formula was mentioned deals with gene fixation instead of rate that genes are replaced.
The K we're calculating is the time for a beneficial mutation to become fixed. K times a number equals 1. That number is the number of generations required to get the frequency of an allele up to 100%. But once you get above 60%, it's not going to take that many more generations before it reaches 100%. I think this answers your objection. Again, we're talking about ONE mutant. It should be quite obvious that more than ONE mutation is required. We're talking about what, a one amino acid change in a protein?PaV
May 8, 2008
May
05
May
8
08
2008
07:38 PM
7
07
38
PM
PDT
of course not. but the whole point of your post was that the selective factor was impossibly high, and thus evolutionary theory could not even begin to explain the phenotypic changes. however, this is only true if you use a population size that is off by an order of magnitude or so from the real data. so your post is incorrect in its implications. agreed?dmso74
May 8, 2008
May
05
May
8
08
2008
07:34 PM
7
07
34
PM
PDT
Something's fishy about this. There's a small sample of the book at: http://books.google.com/books?hl=en&lr=&id=ZUT4zheiNWsC&oi=fnd&pg=PR7&ots=K_LzvERt1S&sig=xvd2OtLWNgW5EziK9902PzdbDaE#PPA27,M1 (Or just google “Theoretical Aspects of Population Genetics”) The page with the formula that you referenced isn't included in the preview (probably on page 11 or 12), but it was mentioned again on page 27. One quote stood out: "From the estimation made so far, K is at least 20" I can't tell how without the full text, but It seems that you've misunderstood something. It doesn't make sense the way you presented it anyway. For example, if we use a very small number for Ne, and say that all mutations are neutral (s1 = 0), then the rate of gene substitution should be similar to the mutation rate(v). But the formula just gives a value of 0. There's a number of alarm bells going off for me on this post. Including the fact that most of the paper around where the formula was mentioned deals with gene fixation instead of rate that genes are replaced.dreamwalker007
May 8, 2008
May
05
May
8
08
2008
07:24 PM
7
07
24
PM
PDT
Yes PaV, much better. Thanks. And dmso74, thanks for answering my other question about average population size.Atom
May 8, 2008
May
05
May
8
08
2008
07:08 PM
7
07
08
PM
PDT
dmso74: Would you also like to say that the enlarged skull and bite, the cecal valves and the changed behavior are the result of ONE mutant?PaV
May 8, 2008
May
05
May
8
08
2008
06:59 PM
6
06
59
PM
PDT
You have the units for K wrong. Check page 11.Secretly
May 8, 2008
May
05
May
8
08
2008
06:18 PM
6
06
18
PM
PDT
i would respond by saying that an improbably high s value is a lot different than an impossible s value.dmso74
May 8, 2008
May
05
May
8
08
2008
05:50 PM
5
05
50
PM
PDT
dmso74..... The formula for s is s = K / [4 x Ne x v]. If 10^-5 is in the denominator, then you can very simply place it in the numerator and switch sign; so it becomes 10^5 in the numerator. Then the formula is s = K x 10^5 / [4 x Ne x 1.5] s = K x 10^5 / [6 x Ne]PaV
May 8, 2008
May
05
May
8
08
2008
05:40 PM
5
05
40
PM
PDT
Sorry, I left out the "4". So, we're looking at s = 0.15-0.19PaV
May 8, 2008
May
05
May
8
08
2008
05:27 PM
5
05
27
PM
PDT
dmso74: First, thank you for picking out the numbers. However, if we assume a doubling of population each generation, let's just say that it takes 8 years to get to an average population size of 4,000. Ne is 3200. Instead of 36 years, we need to use 28 years. The formula is: K = 4 x Ne x s x v v= mutation rate = 1.5 x 10^-5. So, s=1/[28 x 3200 x 1.5 x 10^-5] s=10^5/ 28 x 4800 s=100,000/134,400 s = .75 If you insist on using 5,000, then s = 4/5 x .75 = 0.6 This is an extremely high value for s. Nonetheless, this is what s must be for there to have been just ONE beneficial mutant becoming fixed in the number of years recorded. How do you respond? P.S. Atom, is the equation clear enough now? The 1/36, or the 1/28 numbers represent K. I calculate K reasoning this way. We have ONE mutant fixed. And the denominator represents the number of generations needed to have that ONE mutan fixed. (I'm assuming one breeding season a year.)PaV
May 8, 2008
May
05
May
8
08
2008
05:25 PM
5
05
25
PM
PDT
actually it looks like the real problem is that you mixed up 10^5 and 10^-5 somewhere..dmso74
May 8, 2008
May
05
May
8
08
2008
05:11 PM
5
05
11
PM
PDT
could you please make the following corrections to this post, based on data taken from the paper in question: the original population size was 10 lizards, not 12 the final population size was 5,000 lizards. THis suggests that your estimate of the average population size of 250 is orders of magnitude too low. If you substitute in a more reasonable population size, say 5,000, the math works out perfectly fine. s = 1.39 x 10^-6.dmso74
May 8, 2008
May
05
May
8
08
2008
05:03 PM
5
05
03
PM
PDT
PS By equation formatting I mean putting the equations on separate lines with spacing between each line, like this: y = x ^ 2 2N = 180 * 1/p etcAtom
May 8, 2008
May
05
May
8
08
2008
03:51 PM
3
03
51
PM
PDT
Glad to see you putting numbers to the problem. A couple of suggestions: 1) Perhaps you can do some equation formatting to make it easier to follow your computation? 2) I think people might take issue with your assumption of 250 for average population size. If this estimate is based on some empirical reasoning, you propbably should show why you arrived at that number. Does the original article hint at the current population size?Atom
May 8, 2008
May
05
May
8
08
2008
03:46 PM
3
03
46
PM
PDT
1 2

Leave a Reply