Dear Mark,
Thanks for summarizing what you think my opinion is on the topic of randomness of mutations. After careful scrutiny of your mail I like to repond as follows.
First you say that:
"Unless I am much mistaken, your present & previous arguments are summed below. A to E.
A/ Random mutation isn’t random, so, the tenet that says evolution is random mutation culled by natural selection falsifies the NDT.
I say:
A better representation of my opinion is that mutations are of two types:
1) Random mutations. They are random because they are due to oxidative stress, UV, etc.
2) Non-random mutations. These mutations are non-random in the sense that they are introduced in a particular region of the genome (gene), are mediated by proteins and/or RNA mechanisms, and maybe we are even able to predict where they are introduced. I postulate that these mutations are introduced in response to the environment.
Next you quote Futuyma, who has a different opinion that you cling to:
quote:
--------------------------------------------------------------------------------
Mutation as a Random Process
Mutations occur at random. It is extremely important to understand what this statement does & does not mean. It does not mean that all conceivable mutations are equally likely to occur, because, as we have noted, the developmental foundations for some imaginable transformations do not exist.
I say:
"How does Futuyma know that they do not exist?"
It does not mean that all loci, or regions within a locus, are equally mutable, for geneticists have described differences in mutation rates, at both the phenotypic & molecular levels, among & within loci (Woodruff et al. 1983; Wolf et al. 1989). It does not mean that environmental factors cannot influence mutation rates: ultraviolet & other radiation, as well as various chemical mutagens & poor nutrition, do indeed increase rates of mutation.
Mutation is random in two senses. First, although we may be able to predict the probability that a certain mutation will occur, we cannot predict which of a large number of gene copies will undergo the mutation.
I say:
"This is Futuyma's opinion on the topic. I do not agree."
The spontaneous process of mutation is stochastic rather than deterministic. Second, and more importantly, mutation is random in the sense that the chance that a particular mutation will occur is not influenced by whether or not the organism is in an environment in which that mutation would be advantageous.
I say:
This remains to be establihed. It is increasingly demonstrated that gene rearrangements change in response to the environment. As a recent example I would like you to focus on the Enterococcus faecilis (Nature 2002, July 13th). This microorganism possesses a 150 kb DNA isle that renders insusceptibility to an enormous amount of antibiotics. The authors concluded that E. faecilis is able to shuffle DNA regions and to actively mutate this region (by an unknown mechanism). These are DETERMINISTIC mutations, namely determined in this particular DNA isle.
--------------------------------------------------------------------------------
(Evolutionary Biology 2nd Edition, Douglas Futuyma p281-2)
You say:
"One of the reasons I felt I had to get everything under one roof was the opportunity to reiterate this argument."
I say:
You are entitled to reiterate this 'argument' as often as you like, but I want you to know that it is NO argument. Such arguments are known as "ARGUMENTUM AD VERECUNDIAM", meaning you take an expert's opinion as an argument. However, it is nothing but an opinion! I do not agree with Futuyma's opinion so you cannot use his opinion as an argument. In English it could be translated as AUTHORITIES FALLACY. You encounter them a lot in science: "He says this or that and he is the expert in field SO it is true. Well, I don't think so. The word SO means that what is going to come is logically connected to the previous. Here it is not. SO, it is fallacious (here I use SO properly)."
You have never substantially responded to the issue of the intended definition of random, as outlined above. It is incorrect to insist on a definition that was never intended, & as such, applying a strict statistical definition of random falsifies nothing.
I say:
"The definition is nothing but an opinion."
YET, you still insist you have falsified the NDT because of the statistical non-randomness displayed in mutations. You do this without bringing anything new to the discussion, & I must therefore conclude you are using the time honoured tactic of, I have no argument, so I’ll repeatedly reassert my initial premise. I am giving you the opportunity to bring a new argument to the debate, rather than your reassertions.
I say:
"I don't have to bring anything new to the discussion, since you did not rebut the example of non-random mutation in the 1G5 gene. In fact it is you who is reiterating over and over that I did not falsify NDT (although you already admitted it in a previous mail)."
And you say:
"The intended meaning of random, is that if the locus of a particular mutation cannot be predetermined, it is random (among other things, above)."
I say:
"Even if the gene can be predicted it is NON-RANDOM. Time will learn that genes do respond to environmental change. Time will also reveal the mechanism and then we will be able to do predictions which genes will change in response to distinct environmental factors (e.g antibiotics). The E. faecilis will be a nice model organism."
And you say:
"This means hotspots, & even directed evolution (should such a thing exist), are random."
I say:
"You are continuously repeating that mutations are RANDOM. I already pointed out that there are two types of mutations: RANDOM and NON-RANDOM, DIRECTED, environment induced. So, I don't share your opinion. Furthermore, your above sentence is contradictory. How can something 'directed' be 'random'?"
You say:
"Let us assume that directed evolution exists, & is active in a population of 1,000 individuals. One locus is under potential directed evolution, the other 499 nucleotide sites exhibit statistically random mutational probability. After 1 generation, 250 of the organisms have had the sequence undergo directed mutation, 500 experienced no mutation at all, & 250 experienced statistically random mutation. Could you have predicted the mutation loci deterministically in a given sequence? No, of course not. Ergo, even directed evolution is random (Futuymas definition)."
I say:
"This is all based on your assumptions and Futuyma's opinion. Furthermore, if you sequence all organism you will be able to find the directed mutations, since there will be a fraction of the complete population that have the same mutations. Have a carefull look at the 1G5 gene. Three of the subpopulations have exactly the same mutations, while they are as far apart as Italy, Peru, and USA. Randomness plus selection? No, according to the authors the complete region is in accord with neutral evolution"
You say:
"You cannot gainsay evolutions intended meaning of random. Only the definer/author can do that."
I say:
"So, you agree that it is all a matter of opinions of experts?"
You say:
"Given that you can’t put words in people’s mouths, & tell them what they are supposed to have meant by random, you cannot insert XXX meaning of random in order to falsify random mutation
being required for evolution. It ain’t allowed."
I say:
"By now, you should know what I mean with NON-RANDOM."
You say:
"The next time you say you have falsified the NDT because of randomness, it had better come with a bloody good reason, or I claim victory."
I say:
"Claim Victory?" Victory over an opinion? Yours (Futuyma's) over mine? If that makes you happy, go ahead."
And you say:
"I’m tired of explaining the same thing over & over, & you just reasserting your original position without explanation.
OK?
It’s reassertions like this.
quote:
--------------------------------------------------------------------------------
since I demonstrated that some genes change non-randomly and that falsifies NDT
--------------------------------------------------------------------------------
..that ignore everything that has been said, & it’s beginning to piss me off."
I say:
"You admitted the falsification one mail ago, and now you don't. So, it is you who gives me opposite signals and thus I should be tired, not you.
Next, Item B/
Your summary:
"B/ You believe you have falsified neutral rate mutation/ neutral theory. Therefore phylogenetic analysis cannot be inferred because directed evolution cannot be excluded. Put simply, you are saying that because alleged neutral sequences display non-neutral behaviour, this must be directed mutation, because it’s supposed to be neutral.
Given you accept functional constraint, or at least accept that it’s well established."
I say:
"Better would be that I used a DNA region that behaves according to neutral theory to demonstrate that mutations are not RANDOM in this particular region, and that alone overturns NDT". I did not say anywhere in my mailings that I've overthrown neutral theory. Actually, Neutral Theory it is well established (although it depends on the DNA regions you study). Anyway, what do evolutionists need a neutral evolutionary theory for? If it demonstrates something it is stability of protein/phenotype."
quote:
--------------------------------------------------------------------------------
Ever heard of neutral evolution theory? What does it say for DNA sequences that are not under selective constraint? Indeed, the suppose to change more rapidly!! In fact this has been well established.
--------------------------------------------------------------------------------
Then what do you infer from a nucleotide site that doesn’t display neutral rate substitution? It seems to me that you disregard your acceptance of functional constraint, even though it is a perfectly plausible explanation, in favour of it was designed to mutate there!.
This means that previously functionless sequences may have loci that have function, after all, since functional/selective constraint is observed. Remember, it is YOU who are claiming a falsification, & it is YOUR argument that must therefore have near zero tentativity. Therefore you must SHOW that an ENTIRE sequence is functionless.
I say:
"I have the feeling that you mix up "function" and "neutral rate of evolution". In biology it is accepted that functional --protein-coding-- DNA sequences demonstrate an open reading frame. Such regions can perfectly well change at a neutral rate, demonstrating that there is no selective constraint on the protein."
You say:
"I don’t deny that the original transcribed protein has been ruined. But I do not accept that there is no function at any locus, or never has been, unless you show otherwise. But given that there are pseudogenes/transposons that HAVE been shown to have function, it is imperative that you investigate for this possibility before claiming falsifications."
I say:
"Here you refer to pseudogenes, and in particular to the GLO example.
I don't get your point. The GLO gene is inactivated by a stop codon in exon X, and in the paper the authors claim that this is the reason why primates do not produce vitamin C". If other loci have functions than for sure they do not contribute to synthesis of GLO protein, since it is absent in all great apes. The degeneration is a perfect example of what happens to redundant genes (and the GLO protein is redundant --> see a previous letter) over time: degradation, information to synthesise Vit C more efficiently is lost. It is de-evolution. In fact, if we had had the present knowledge on vit C synthesis before, we had been able to predict which genes might be inactivated over time. In general, redundant genes are supposed to be inactivated over time, since they are not under selective constraint.
You say:
"If you cannot do this, then you cannot infer falsification from non-neutral rate loci in alleged functionless sequences."
I say:
"All I claim is that if a non-functional gene demonstrates a NON-RANDOM position you cannot claim it as proof for common descent, since it might be due to a mechanism. And, if a mechanism is involved the other mutations might as well be introduced similarly."
My comments on item C:
"C/ You assert it is possible to locate hotspots in ancient sequences, in order to be able to infer that pseudogenes (GLO) were wrecked by hotspot mutation.
Firstly, I doubt this very much, since GLO genes are active in more mammals than not. If a hotspot destroyed our GLO function, then why not a cows? Given it has hotspots too.
Secondly, why would an ID design a functioning gene with a built in self destruct? It's like building a plane with wings that fall off, or a car where the steering wheel comes off in your hand above a certain speed."
I say:
"'Why-questions' are metaphysics, not science. But, maybe it is some form of de-evolution, or some form of stability that is acquired in response to the environment (in cows)"
You say:
But I digress.
I agree, you can take an extant sequence, mutate it a thousand times, then examine where the sequences mutate most in order to demonstrate where the hot spots are.
But, this is entirely different a proposition from being able to infer hotspots from ancestral sequences. The best you can see from an ancestral sequence is that a loci mutated a maximum of three times, that is, from A to G, A to T, A to C, certainly not from within your paradigm. There is no way you can quantify a hotspot that has a thousand-fold higher probability of mutation than a statistically random probability. You may look at loci that appear to be random to you, but have mutated thousands of times more than adjacent loci, but because it started as A, & finished as A, you will never know.
This is probably best illustrated in the form of a question;
Q/ There is a single homologous locus in a gene in eight different related organisms.
1/The nucleotides are A,G,T,C,A,G,T, & C. Is the locus a hotspot?
2/The nucleotides are A,A,A,A,A,A,A, & A. Is the locus a hotspot?
3/The nucleotides are A,A,A,A,A,A,A, & T. Is the locus a hotspot?
So, how can you tell the location of hotspots in ancestral DNA?
I say:
"By comparison of subpopulations.
To stick to your example. In subpopulations we find the following sequences:
sequence 1: tattgattagtgg
sequence 2: -------------
sequence 3: --c----------
sequence 4: -----g---a---
sequence 5: -------------
sequence 6: -------------
sequence 7: --a--ag--a-a-
sequence 8: -------------
sequence 9: --a--gg--a-a-
sequence10: -------------
sequence11: -------------
sequence12: --a--tg--a-a-
sequence13: -----------a-
[- = same nucleotide as sequence 1]
Furthermore, if it is demonstrated that the sequence is not under selective constraint my conclusion would be that in this particular DNA segment the mutations observed in the subpopulations are introduced NON-RANDOMLY. Why? Because the same spots and same nucleotides are involved (compare this with the 1G5 gene).
My comments on D:
You say:
"D/ You claim to have falsified natural selection, specifically."
I say:
"No, I demonstrated that that there examples of protein-coding genes that do question the validity of natural selection as being responsible for the residence of these genes in the genome. For instance, the a-actinin genes, and in fact all other redundant genes."
You say:
"Natural selection has been demonstrated umpteen times experimentally.
I'll use a single example, the guppy, Poecilia reticulata. In waters populated by the predator Crenicichla, males have smaller less conspicuous spots that match the gravel bottom (different bottoms elicit different patterns). In effect the guppy has evolved camouflage. The alleles that express phenotypes are under SELECTIVE pressure.
Guppies that exist in waters that lack Crenicichla display a much wider range of colouration. That is to say the alleles that affect skin colour are no longer under selective pressure.
Guppy populations that are in waters that have Crenicichla populations, & are placed in waters without the predator soon display a wider variety of colouration. Again, the skin colouration alleles aren't selectively constrained, & are able to increase via genetic drift, since they are now "neutral" alleles.
If guppies from non-predatorial waters are placed in water with Crenicichla, the colourations soon begin to match the gravel bottom. That is, alleles responsible for skin colouration are under selective pressure.
(Endler 1980)
Now, given you have "falsified" natural selection, can you explain the above within your shining new paradigm?"
I say:
"Where does this example help evolution? Nothing was created here by evolutionary forces. All your example demonstrates is that PREEXISTING alleles may vary in frequency in response to the environment. This is the role of natural selection: choose from alleles that are already present in the population, so the organisms do not become extinct immediately.
You say:
"I don't think so, there is no other explanation other than non effectively camouflaged guppies end up as lunch for Crenicichla, those that have camouflage go on to repopulate, taking their existing/newly appeared alleles with them. Namely, natural selection. If allele frequencies are being affected by natural selection, & natural selection is part of the NDT, then how can you possibly conclude that the NDT can't be seen to work at the genome level?
Falsification of natural selection? Not."
I say:
"Genetic redundancies demonstrate that natural selection cannot be the only mechanism responsible for the maintenance of the genome"
You say:
I saved the best ‘till last
In response to E:
you say:
"E/ You assert that statistical non-randomness as evidence of design."
NON-RANDOM mutations indicate that all information is already present in organisms. Why? Because, the proteins and/or RNA that directs this are specified in the genome and it is reasonable to assume that no selective constraint is on these proteins, since they are only induced in response to the environment. If no selective constraint is involved it requires another mechanism to maintain in the genome as a functional gene. It is my personal opinion that the major part of the genome contributes to these mechanisms. It should be realized that only the minor part of the 40000 genes of man are of known function. Yes, only a couple of thousands have been described (5% or so), the rest is just there. Function unknown. I postulate that these proteins may be involved in maintenance of genes and stabilize the genome. In addition, these proteins may be involved in directing mutations towards the genes in response to the environment. Time will learn.
However, since all these genes are around without selective constraint (e.g. 98 % of the genes of Arabidopsis is redundant) it clearly demonstrates that there is a creator who designed them (for later use).
You say:
"How can you tell a naturally occurring system/object from a designed one?"
I say:
"By redundancies: convenient to have and ready to use if needed. The genome anticipates on changes"
You say:
"Since you are attempting to show ID, then if you can’t answer this question, then ALL your arguments come to nought."
I say:
"In fact I was not attempting to demonstrate ID (but rather I registered to falsify NDT), but anyway, now I did."
You say:
"Retrospectively, your problem...."
I say:
"I didn't know I had a problem."
"...is your own use of language. You drop the word "falsification" in to your argument, without realizing it requires you to have all bases covered. You simply have to have a lot more information before you can get anything like a falsification of the NDT.
I say:
"Do you wanna define what is understood by a falsification? It will turn out that the 1G5 gene falsifies NDT"
Best wishes,
Peter
[This message has been edited by peter borger, 08-22-2002]
[This message has been edited by peter borger, 08-22-2002]
™ Version 4.2