Beneficial mutations like the ones that cause sickle cell anemia actually deteriorate the information in the genome by deforming the red blood cells and making them less, not more, efficient.
This basically means nothing. You can't quantify the net change in information involved. All you can do is make a subjective judgement and then label it a loss of information. Does the change in haemoglobin make it a less efficient oxygen binder, no. Does it change the efficiency of the blood to transport oxygen, yes. Does it increase the carrier's resistance to malaria, yes. These tell us some of the things that the mutation to the sickle cell from of haemoglobin (HBsc) leads to but it doesn't give us a way to quantitate the informational change involved.
Becasue you always have to remember that all individuals have mutations. The selection does not happen between mutated and non-mutated individuals. But between less mutated and more mutated.
It would be more accurate to say that the selection is between differently mutated individuals, rates of mutation in populations are relatively consistent. It is the nature of the specific mutations that are important, not simply the number of them.
And as time goes by, mutations accumulate, and lead to the genetic meltdown.
Maybe, hypothetically, or not as the case may be. As we have discussed before there are plenty of models which do not lead us to expect genetic meltdown to be an inevitability, including models with synergystic epistatic interactions between mutations and those incorporating nearly neutral balancing mutations.
Is genomic 'information' for you simply a function of organismal fitness? Something divorced from the actual specifics of the underlying genetic change. Is it tied to the specific biochemical functionality of the individual proteins? If so who determines what functional changes represent increases or decreases in information? We've touched on these questions before but you have never answered them satisfactorily.
That's like saying that a sequence that codes for an eye is replaced with a sequence that codes for nothing, is not a net change.
Yes, it is like saying that because anyone who talks about 'a sequence that codes for an eye' is essentially saying 'I don't understand developmental biology' in the same way your statement shows that you don't have a coherent understanding of genetic information.
Of course it does affect it's oxygen binding ability, that is one of the points.
It affects it when the HBs forms polymers, but the actual individual haemoglobin molecules retain their normal binding affinity (Bonaventure et al., 1999.
The resistance itself is not a biological function. It is an inability of malaria to infect the body.
This is what I'm talking about, you make a totally subjective judgement call that undercuts the validity of your whole approach. You are totally divorcing the concept of biological information from what we identify inc complex evolved systems, it isn't a question simply of having some form of molecular function, it is about wider biological function in terms onf environmental interactions, which resistance to amlaria would certainly qualify as.
As to looking at nucleotide changes to quantify the loss, this is no answer at all, you could have just as many changes with no effect on function whatsoever.
This link here explains how 110 mamalian species were tested and a clear case of accumulation of slightly deleterious mutations was noticed.
I'm quite happy to accept that the mitochondrial genomes of larger mammals with small effective population sizes may be subject to genetic meltdown. What I dispute, and this paper does nothing to support, is the contention that this is a universal trend in all genomes, or even all vertebrate genomes. The paper certainly doesn't demonstrate the inevitability of genetic entropy leading to extinction.
And as you can see from this here graph, it doesn't stop the accumulation.
I see that you still don't understand that graph, it doesn't say anything about changing the accumulation of mutations, it talks about the difference epistatic effects have on the fitness effects of cumulative mutations. The point is that the steep drop off in the negative synergistic epistasis line means that the accumulation of only a few mutations will produce a non-linear change in fitness and consequently a non-linear change in selective pressure against the carrier.
Genetic informations are the biological function that is encoded in the genome. For an example, if you have a sequence that codes for an ATP synthase, and you mutate it so that after a while the same sequence produces a machine that does not produce energy anymore, and simply stands there, it lost information. If it so happened that in acquired new information for producing something else, than that would be a gain in information.
This is the sort of vague definition I am talking about. This doesn't tell us how to measure information at all. All it does is let us identify the extreme case where the gene completely loses its function. Have you settled on the Hazen/Szostak form of 'functional information', if so you should bear in mind the limitations they themselves identify with their approach.
I already did. I gave you two examples. And now I'll give you a third one. As you can see, bacterial resistance is also gained by loss of efficiency. Now, it's your turn. Show me some beneficial random mutations that work wonders...
quote:... [A] microorganism can sometimes acquire resistance to an antibiotic through a random substitution of a single nucleotide... Streptomycin, which was discovered by Selman Waksman and Albert Schatz and first reported in 1944, is an antibiotic against which bacteria can acquire resistance in this way. But although the mutation they undergo in the process is beneficial to the microorganism in the presence of streptomycin, it cannot serve as a prototype for the kind of mutations needed by NDT [Neo-Darwinian Theory]. The type of mutation that grants resistance to streptomycin is manifest in the ribosome and degrades its molecular match with the antibiotic molecule.
Once again you use an example which only someone with no idea what they are talking about would use to show a loss of efficiency. Binding to streptomycin is not a function of the ribosome, binding to the ribosome is rather a function of streptomycin. So it is not necessarily a loss of efficiency for the ribosome to change its structure to reduce the binding affinity to streptomycin. You have to use a crazy sort of logic to portray a mutation which has no effect other than to reduce the affinity between the ribosome and streptomycin as a loss of function/efficiency for the ribosome.
The reason this is such a dumb example is that there are perfectly good examples of streptomycin resistance mutations which have genuinely deleterious side effects, including the production of strains which were essentially streptomycin dependent (Springer et al., 2001).
Also causing the reduction of information, becasue the the reduction of specificity is reduction of one component of complex SPECIFIED information.
Only if you use your own crazy version of it, rather than one which has any relevance to actual biological function of the changing molecule! Alowing that CSI is represented in this system the specification surely resides with the genetic sequences allowing the production of Streptomycin with a structure allowing it to bind to ribosomal elements, not the other way around. You might argue that the change in the ribosomal element effectively reduces the information content of the streptomycin biosynthesis sequences, but I don't see how you can argue that it represents a loss of information for the ribosomal sequence.
My example is better because it relates to actual deleterious functional effects to the ribosome, which the various informational metrics involving functionality we have discussed would probably all identify as a loss of functional information. Of course there is a counter-argument that we need to be able to calculate the potential gain of functional information that the resistance phenotype represents in order to determine if the informational change has led to a net loss or gain of information.
While you are obviously correct to bring up the difference between the traditional mendelian concept of recessive genes in sexual poulations in contrast to an asexual haploid organism I think there is a case to be made for potential forms of recessive inheritance in bacteria.
The obvious one is the presence of multiple forms of genes in plasmids outside of the central bacterial genome. This could easily allow for the presence of a gene whose phenotype was suppressed in the presence of other 'allelic' forms of the gene.
The other possibility is for multiple copies of a particular gene to be present in the bacterial chromosome. When similarly a particular 'allelic' form may have its phenotypic effect masked.
Clearly these 'recessive' form scenarios have major problems, principally how such effectively functionless copies would be preserved in the population waiting for an evironment where their function could finally be realised. Essentially positing such dormant 'recessive' genes makes an extremely unlikely hopeful monster scenario. Arguably many antibiotic selection experiments rely on replication in a non-selective environment to generate a large panel of 'hopeful monster' variants, but we know from experience that the levels of bacterial multiplication and mutation rates allow us to generate enough 'hopeful monsters' for it to be a viable strategy.
In a less extremely selective environment we don't need to rely on such 'hopeful monsters' pre-existing as they can arise within an ongoing population, this would perhaps be better modelled by a bacteriostatic environment where the antibiotic only limits or strongly reduces the growth rate of the population rather than actually killing it. Since Streptomycin has a bacteriostatic effect at lower concentrations I'm not sure which scenario would more accurately reflect the conditions under which streptomycin resitant variants usually arise.
What got changed, the ribosome or the streptomycin? Obviously the ribosome. Therefore, it is the one that lost specificty.
Except that sequence wasn't evolved to have specificity to streptomycin. If I shape my hand to fit into a particular hand hold when climbing a rock have I suddenly increased the rock's 'information' because I have specified my hand to fit it? You seem to be saying that as soon as the full streptomycin biosynthesis pathway had evolved the specified information content of the ribosomal genes jumped despite there being no change in its sequence, we might consider that information to have been 'free' since it only served the function of the antibiotic rather than the ribosome. I don't see why this is more logical than me suggesting that it can lose that binding specificity without losing information content. All you seem to be doing is highlighting why your conflation of the specificity in CSI and binding specificity is meaningless when you don't actuslly look at the functional biological context.
Once again you are making up a system where things can't help but 'degenerate' because you are defining any change from the initial state as degeneration. Would you consider any mutation which increased binding affinity to anything to therefore represent an increase in information/CSI? Even if the subsequent binding served no functional purpose for the mutated protein?
You are just showing us why these creationist/ID measures of information divorced from actual considerations of function are totally useless. They are just ad hoc measures based on arbitrarily selected starting points and in many cases arbitrary criteria for what constitutes a gain or loss of information.
Remember the Durston et. al FSC measure we discussed, where they base their estimates essentially on conservation, totally regardless of actual functionality beyond very crude classification. Their method means that any novel mutation which allows the maintenance or even improvement of function for a protein nevertheless would represent a loss of FSC. There is no possible route in their approach which allows a novel mutation to produce an increase in functional information.
The number of times you say it, doesn't matter. Genetic entropy IS a cumulative loss of fitness, leading to mutational meltdown and extinction.
This is such an important point. Unless SO is talking about a totally different concept of Genetic Entropy from Sanford then it must absolutely be about actually deleterious mutations and decreases in fitness. Abstruse notions of beneficial mutations which reduce some unmeasurable notion of information are totally beside the point.
The second mechanism is negative selection. That is, those with the highest deleterious mutation rate are selected against. This ongoing process stops Muller's ratchet.
Another important element in this is the existence of synergystic epistatic effects, where the fitness burden of multiple deleterious mutations is greater than the product of their effects in isolation.
On the 649th post of this thread you refer to a link that may or may not be on this thread. Message 115. The link does not appear to be what you claim it to be however, why exactly do you think this paper has any impact at all on evolution?
Lets be explicit here. The paper is specifically only about certain mammalian mitochondrial genomes a very small and discrete genetic subset distinct from the chromosomal genome. It further does not show that these 110 species are doomed to genetic meltdown and extinction, it merely suggests that as mammals grow larger their effective population size tends to reduce and they are therefore more prone to the accumulation of slightly deleterious mutations, as we would expect given the effects of small effective populations on drift.
The paper says some interesting things about the effects of increasing body size on population genetics and possible consequences in terms of trends in extinction. It certainly doesn't provide any evidence that all species are on a continual downward spiral of genetic entropy which will inevitably end in genetic meltdown.
You can't make it sensible to talk about binding sites which have evolved/arisen to bind a particular protein representing CSI in the binding target in all cases. Using your logic every time an antibody is raised to a different epitope on a protein the information content should rise! Every animal with an adaptive immune system is increasing the genetic CSI content all the time!
But now you talk about 'original information content' which is quite different, I put it to you that the 'original information content' would have arisen before the full streptomycin biosynthesis pathway, so in fact all you are losing is ,as I suggested before, the 'free' informational value imparted by the development of streptomycin biosynthesis rather than any of the 'original information content'.
But what you really seem to be saying is, once again, that any change from the first sequence derived from a gene is a loss of information, or in Durston et al.'s approach essentially any deviation from the consensus from an alignment of related sequences.
The binding specificity whatever it is for, seems totally unrelated to what you are saying, you are trying to hang some element of functionality on it when there simply isn't any. The only functional effect the mutant has is to allow the bacteria to survive in the presence of streptomycin. It hasn't lost the function of binding to streptomycin becuase that was never its function.
The ribosome is arbitrary in some ways, it is not 100% conserved amongst all species so clearly there is some allowable variation at different positions. You have decided to arbitrarily decree that any mutation changing the sequence from its inital state is a loss of information. Similarly Durston et al. arbitrarily decree that any change away from their consensus sequence will be a loss of information.
You talk about multiple mutations giving rise to a functional ATP synthase gene, but that misses my point about novel mutations. If you take a random sequence and put it through multiple rounds of mutation and selection until you eventually produce a sequence which matches that of a consensus ATP synthase then using Durston et al.'s method we will have arguably increased the information for that sequence, but only to match a sequence we already had we haven't generated truly novel information.
Durston et al.'s method won't let us measure the information we have created if we experimentally evolve an ATP synthase enzyme with a radically different underlying sequence, all that would do in fact if we added our new enzyme to the alignment would be to reduce the functional specified complexity for the whole gene family of existing ATP synthases, ormore probably the program would fail to return a meaningful result because the sequences wouldn't align at all.
Similarly if we were to produce an ATP synthase with a truly novel mutation, i.e. one not extant in the family of functionally related genes, which improved the rate of ATP synthesis Durston et al.'s method would again tell us that our new sequence has less functional information than the consensus sequence and would again reduce the FSC for the whole family of ATP synthase genes.
My point was that you Durston et al.'s approach sets an arbitrary maxima for FSD based on the consensus sequence. It will not allow you to measure an increase in FSC even if a novel beneficial mutation arises.
In contrast Haze et al.'s method uses actual measures of functionality as an important element so it would allow you to quantify an informational increase based on such mutations.
Again we see how the IDist route of ignoring actual details of biological function for vague proxies makes their approach fruitless.
You missed the entire point. You claim that the specific binding between streptomycin and the ribosome is due to CSI in the ribosomal sequences. So why is specific binding between antibodies and the epitopes they bind not due to CSI in the sequences which code for that epitope? Did you quickly evaluate the bits required for all possible epitopes and determine that they were all under 400 bits?
The 'free' information, as I have pointed out more than once now, is the information which has to have suddenly magically been added to the coding sequence for the ribosome when the full streptomycin synthesis pathway had developed and it suddenly had a high affinity binding partner.
You say you don't claim mutations are necessarily a loss of information, but you persist with your illogical claim that it is a loss of information in the streptomycin resistance example. Can we just agree that its not a net change in information? Would you go that far?
To claim that any change in structure is a 'degradation' is bringing front and centre your assumption that the initial structure is some sort of optimal or ideal structure. It is totally inconsequential to the effect of any particular mutation that if you keep on changing the structure long enough at random its function will disappear.
Nice to see your basic understanding of entropy is as faulty as your concept of genetic entropy. You have, as usual, conflated any specific change with the average tendency of all random changes.
The fact that in general non-neutral mutations will tend to have a deleterious effect on a proteins function doesn't mean that beneficial mutations don't exist. You do understand that entropy is a statistical phenomenon?
Going from the general to the specific as you do totally defeats the entire point of this discussion.
Re: Genetic Entropy[quote]On the 649th post of this thread you refer to a link that m
Yes it does because most bad mutations are fatal.
If by 'bad' you mean the same thing as deleterious then I would suggest you are wrong. There is certainly plenty of scope for lethal mutations, or mutations leading to reproductive sterility, but there is an even greater spectrum of more mildly deleterious effects as a result of mutation. There is also the fact that many mutations which may be fatal in homozygotes are merely detrimental, or even neutral, in heterozygotes.
Most studies suggest that the highest proportion of mutations lie in the 'nearly neutral' region, where they only have small beneficial or deleterious effects.
Certainly these studies may underestimate the spontaneous rate of embryonic lethal mutations if they are performed on an extant populations genetic makeup, but there doesn't seem to be any reason to suppose that most deleterious mutations are so severe as to be lethal. I might even go so far as to suggest that given the robust nature of many genetic networks this may even be true for entire gene deletions/ nulls, though I admit I don't have any solid numbers to back that up.
So you don't have time for debate but you have time to discuss your uninformed intuitions with similar minded people? To what end? So you can all feel good about how you agree with each other and Stephen Meyer? So that by sheer weight of accord you can suddenly make all of the evidence contrary to your preconceptions disappear?
Nothing in your post seems to have even a passing resemblance to evidence. Not only are you relying on second hand figures given to you by Smooth Operator, but you seem convinced that intuition and common sense are superior substitutes for actual knowledge and research.
The fact that you can only think of one example of a beneficial mutation just goes to show that you are only familiar with them from web sites rather than the actual scientific literature, indeeed even just reading EvC forum it shouldn't take you more than a few searches to turn up many further examples of beneficial mutations outwith bacteria, i.e. mutations protective from atherosclerosis, HIV infection, pesticide resistance.Or is this yet another ID/Creationist version of 'beneficial' which is unrelated to simply confering a survival/reproductive benefit on the organism?
As you yourself concede subsequently pretty much the entire rest of your evidence is simply appealing to the authority of Meyer and Behe. You say that we can't counter the evidence itself and since you haven't actually presented any you are correct.
I have to say though the point that the actual amount of 'functional' information compared to the information for the whole protein is a small proportion seems to my mind to be an argument against an ID interpretation, it means that as has often been pointed out, all of those calculations of information/CSI/functional information based on the full length of a protein/DNA sequence have been inflated. It is worthwhile noting that protein binding sites do not need to be re-evolved to appear in diverse protein families, there is plenty of scope for swapping such sites between genes allowing for novel recombinations of binding sites and other active sites. So in theory each type of site need only evolve de novo once.
To restrict yourself only to conversing with Brad H who seems to be from exactly the same branch of intuition/web based understanding (No chromosomal mutations in bacteria?) , seems to be a way of cutting out any dissenting voice that might introduce actual factual infromation into your discussion. Why do you think this is a good thing to do?
At the moment you don't seem to be discussing 'What exactly is ID?' but rather giving a practical demonstration of why ID isn't considered a credible science.
Re: addition, subtraction, addition, subtraction, where does it end?
However what I am stating is that I have never heard of a phenotype changing as a result of "observed" added (new) information to the chromosomal DNA.
Then you should look at Richard Lenski's ongoing long term evolution experiment in E. coli. They have used a plasmid free strain of E. Coli and grown them over 20,000 generations. They have identified multiple beneficial phenotypic changes and begun the task of associating them with specific genetic mutations.
One recent example is in the Stanek et al.(2009) paper in which they identify a mutation which confers a 5% fitness increase when it is introduced into the ancestral strain.
Whether this constitutes an increase in information or not depends on how you define, or more importantly measure, information. I'm sure Smooth Operator would classify this as a degradation of information as the mutation that was identified, which is an insertion upstream of a protein coding sequence, is predicted to lead to a reduction in binding affinity for a transcription factor and a subsequent reduction in the expression level of the coding sequence that the mutation sits in the upstream region of.
I understand your, bacteria aren't animals/multicellular organisms, objection but in this case elements like plasmids and conjugation have been excluded by experimental design, so it is a much closer analogue to the situation for an asexually reproducing animal, albeit with a much shorter generation time.
I'm sure if we scrutinised the many lab experiments invovling insect evolving pesticide resistance we would find some interesting candidates to identify just what does constitute an increase in information. Would a gene duplication conferring pesticide resistance be sufficient (this is purely hypothetical, I have no idea if there is a pesticide resistance mutatn of this type)?