Uncommon Descent Serving The Intelligent Design Community

Media Mum about Deranged Darwinist Gunman

Share
Facebook
Twitter
LinkedIn
Flipboard
Print
Email

John West of the Discovery Institute Reports:

But when a gunman inspired by Darwinism takes hostages at the offices of the Discovery Channel, reporters seem curiously uninterested in fully disclosing the criminal’s own self-described motivations. Most of yesterday’s media reports about hostage-taker James Lee dutifully reported Lee’s eco-extremism and his pathological hatred for humanity. But they also suppressed any mention of Lee’s explicit appeals to Darwin and Malthus as the intellectual foundations for his views. At least, I could find no references to Lee’s Darwinian motivations in the accounts I read by the New York Times, the Los Angeles Times, the Washington Post, ABC, CNN, and MSNBC.

Major Media Spike Discovery

Comments
gpuccio, I'm back from my travels and glad to see the discussion is still ongoing. It seems like a good point to summarize where we've reached agreement and what points are still under contention. We seem to agree that evolutionary mechanisms, particularly mutation and differential reproductive success (which results in what is sometimes too loosely called "natural selection") can generate "functional information" (more on my reason for using scare quotes will follow). We also seem to agree that the amount of "functional information" created by evolutionary mechanisms is additive. That is, the "functional information" created by a series of mutations that become fixed in a population is equal to the sum of the "functional information" of each mutation. From the above two points of agreement, we seem to agree that evolutionary mechanisms could, in principle, create sufficient "functional information" to cross the boundary to "Complex Specified Information". I recognize that you do not believe this to be possible in practice, but mathematically there is nothing preventing it. Another point of apparent agreement is that any calculation of "functional information" must take into account the history of the changes between the initial state and the final state of the system being measured. You yourself made this point in the discussion of the evolution of citrate digestion in Lenski's experiment. Even though citrate digestion is a completely new function and therefore, according to your original definition of CSI, meets the criteria of specification, you noted (correctly, in my view) that we should only measure the "functional information" in the mutations that created the function, not for all the components of the genome that support it. This does, however, have consequences for some of your previous CSI calculations. When you have come up with large numbers for CSI of certain proteins, you have not taken into consideration their evolutionary history. Simply computing four to the power of the length of the genome or twenty to the power of the number of amino acids in the protein is mathematically equivalent to asserting that those biological components appeared complete and de novo. I believe we both agree that such an assertion is not aligned with emprical observations of real biological systems. No biologist claims that such structures arise instantly, so demonstrating that it is unlikely that they could do so does not pose a problem for modern evolutionary theory. To keep this of manageable length, I'll discuss what I see as our currently open issues in a separate post. I am interested to know if you agree with me on our points of agreement thus far.MathGrrl
September 10, 2010
September
09
Sep
10
10
2010
09:27 AM
9
09
27
AM
PDT
So, gpuccio, I will now stop to throw citations at you about topics I barely understand. ;-) In any case I think I have established a few things: 1. In principle, evolution can add information to the genome. I think you knew and accepted this from the beginning but it is still good to have established this again, because some ID guys think this is impossible. 2. Even very small changes to the genome can result in a completely different protein folds, something you thought would be impossible. 3. Even if I accept your number of 6000 (which I don´t) the Durston paper is wrong because it doesn´t take these additional ways to improve the fitness into account. In any case his approach is of no use to determine the evolvability of anything, since he doesn´t take evolutionary mechanisms into account at all. 4. There is good science being done on the origin of protein domains. While this is a very difficult topic the examples that are discussed clearly demonstrate that your claim that evolution in principle can not cross from one domain into another is false. Have a nice weekend!Indium
September 10, 2010
September
09
Sep
10
10
2010
08:51 AM
8
08
51
AM
PDT
gpuccio, I think you underestimate the importance of the secondary structure: Such a substantial change as observed in the article I linked is always also resulting in a change of the tertiary structure. So even your new demand has been met. How did the protein domains evolve? That is a very hard question indeed and most of this happened in the extremely distant past. There is some work being done, however. It has been found, that identical sequences for different conditions of the solution can lead to completely distinct 3D structures. http://www.jbc.org/content/277/20/17863 It has been shown that in principle a change in 3D structure can occur without loss of function. Link 1 Link 2 People even write overviews about different scenarios for the evolution of new protein domains: Link 3 Link 4 So, as expected the is no fundamental reason why evolution should not be able to reach different families of 3D configurations of proteins. Regarding Durston: I don´t know how you end up with only 6000 ways to improve the fitness of any given organism by changing its genome. Could you elaborate?Indium
September 10, 2010
September
09
Sep
10
10
2010
08:26 AM
8
08
26
AM
PDT
gpuccio@125:
As for your new example: transitions between beta-strand and alpha-helical conformations Again you are not a biologist and you cannot know. A “transition between beta-strand and alpha-helical conformation” is not a change of folding, but just a change in local secondary structure. No harm done, anyway.
I'm sorry, but this is just wrong. A change in "local secondary structure", from beta-strand to alpha helix, is by definition a change in a structural fold. And it most certainly causes a dramatic leap in overall three-dimensional structure. On this point, Indium is spot-on correct. For the record, I am a biochemist who works on matters related to structure and function of proteins.Arthur Hunt
September 10, 2010
September
09
Sep
10
10
2010
05:47 AM
5
05
47
AM
PDT
Indium: english is not my first language too. I am italian. As for your new example: transitions between beta-strand and alpha-helical conformations Again you are not a biologist and you cannot know. A "transition between beta-strand and alpha-helical conformation" is not a change of folding, but just a change in local secondary structure. No harm done, anyway. Regarding Durston, what he has done is extremely valuable: he has measured functional information in different protein families in a reliable way. Nobody else had done that so brilliantly before. And you are wrong again when you say: Nobody thinks that as a general rule new configurations poof into existence from random sequences. As I have shown, if a new domain emerges from a pre-exixting unrelated domain, the starting state if de facto random in relation to the new funtion which will emerge, because it has no information about that function, it ha s a different folding and is unrelated at the primary structure level (less than 10% homology). Now, according to what we know from natural history, those transitions must have taken place some way. If darwinian mechanisms are not able to explain them, other explanations must be offered. We in ID are offering a very good one. I believe there are 10^35 possible positive reconfigurations which can be reached by Darwinian processes. Again, I will not force your beliefs. Beliefs are personal. Bit what a pity that all the process of evolution, in 4 billion years, and whatever its causal mechanisms may be (darwinian or design related), has found only 6000 of them. Well, at least we have great expectations for the future! :)gpuccio
September 10, 2010
September
09
Sep
10
10
2010
05:16 AM
5
05
16
AM
PDT
Maybe you will never follow, but I have explained many times, especially to you, that artificial selection is a form of design, and that it is completely different, and vastly more powerful, than natural selection.
In the same way that laboratory chemistry is more powerful than "natural" chemistry. I think I understand.Petrushka
September 10, 2010
September
09
Sep
10
10
2010
04:55 AM
4
04
55
AM
PDT
BTW, thanks for your patience so far. English is not even my first language and in addition the topic at hand is complicated enough to make articulation of my thoughts difficult at times.Indium
September 10, 2010
September
09
Sep
10
10
2010
04:11 AM
4
04
11
AM
PDT
gpuccio, so, macroevolution in your sense of the word can be observed when the evolution of a new protein folding is observed? You seem to think thats impossible to generate with a small change to the genome. Again I am not a biologist, but a quick search turns up quite a few articles. Please have a look at this Overview
In addition, in a few cases significant transitions in structure have been demonstrated following one or a few amino acid mutations in a protein sequence. Examples include transitions between beta-strand and alpha-helical conformations in mutants of the Arc repressor [7] and in the Kazal-type serine protease inhibitor domain [8].
Reference 7 and 8 can be found in the linked article. So, your demand for a case where a simple change in the genome can lead to completely different folding types can be met. Now, for your statistics and Durston. I can see two problems there. First of all, I don´t believe your number of 10, 100 or 1000 possible positive changes to the genome. I believe there are 10^35 possible positive reconfigurations which can be reached by Darwinian processes. Now both of us have guessed a number. How can we decide who is right? Until we have come to such a decision, is it correct to take just *one* function into account, like Durston does? Secondly, his calculations are not usefull anyway: Nobody thinks that as a general rule new configurations poof into existence from random sequences. All he really does is putting some numbers on the tornado in the junkyard scenario, neglecting evolutionary mechanisms.Indium
September 10, 2010
September
09
Sep
10
10
2010
04:08 AM
4
04
08
AM
PDT
Petrushka: Maybe you will never follow, but I have explained many times, especially to you, that artificial selection is a form of design, and that it is completely different, and vastly more powerful, than natural selection. If you don't follow, or just don't agree, let's leave it to that. But please remember, in the future, that that is my conviction, and therefore it's useless that you quote to me new papers describimg the potentialities of artificial selection. I agree with that.gpuccio
September 10, 2010
September
09
Sep
10
10
2010
03:25 AM
3
03
25
AM
PDT
Indium: I am happy that we can go on in our conversation in a spirit of serene confrontation: that's all I expect from my interlocutors. As you say that you are not a biologist, I may perhaps clarify some further aspects for you (after all, I am a medical doctor, which could be considered as a "lessere form" of biologist :) ). I will start form the article you quote. The fact is that in that example a single mutation brings a chane in the biochemical activity of the molecule. That's again a very good example of microevolution. As you can see if you look at the paper, especially at figure 5, the mutation certainly modifies the active site, but the 3D structure of the molecule, and its fold, remain almost the same (you can see the slight variations in different colors in the figure, orange and blue). That's why I have kept mt discussion at the level of protein domains and isolated superfamilies. Each superfamily may have lots of different proteins with different functions, sometimes slightly different, sometimes very different, but the general fold and the "general function" are the same. Take, for instance, the case of nylonase, which derives form the penicillinase domain, through, probably, a couple of mutations. The biological function of nylonase is very different form that of penicillinases: the first has a digestive role for nutrition, the second aerves to protect bacteria from penicillins producted by other bacteria. But, biochemically, both are esterases, and they share exactly the same fold and the same biochemical function. The small variation at the level of the active site modifies the affinity for specific substrates (nylon or penicillin), but the enzynatic activity is anyway of the same biochemical type. And the structure is the same. We jhave to distinguish between the general fold of a domain, which usually defines its general function, and the specific active site, usually determioned by a much smaller number of AAs. Small variations in the active site can mokdify substantially the final effect of the protein in a biological context, but they don't modify substantially the folding and general biochemical characterization of the protein. That's why the paper you quote is good and interesting, but again is only a description of a microevolutionary event, and in no way it shows any progression toward a new, different, isolated fold or superfamily, which was my example and context. You say: My point is that evolution routinely finds *many* new ways to improve the fitness of organisms in different situations. From your arguments, I think that you are not a statistician, too. No problem with that. Let's clarify. Let's suppose that in a system 10 different solution which can improve fitness are potentially available, and that each of these solutions has a probability to be found, through ramdom search, of 10^-45. The probability of finding at least one ppf the solution should be approximately (I am not being necessarily precise here, if there is any statistician out there, he can correct me): (10^-45)*10 that is, 10^-44. Which is an improvement of only one order of magnitude, and not a great consolation. To significantly improve the probability, you need "at least", say, 10^10 different functional solutions for that context (which I believe absolutely non realistic), and that would anyway leave the probability at 10^-35, which is not a joke at all. Please note that, at cthe level I have suggested of fundamental functionality (protein domains with a less than 10% homology isolation) we are aware at present of only about 6000 (6*10^3) different superfamilies/families in the global proteome. That means that evolution in all its history, has only found that number of fundamental protein structures. Moreover, the rate of appearance of new protein domains at that level has constantly decreased in natural history. Do you really believe that in the search space 10^10 or more functional structures still lie undiscovered, neither by nature nor by us? Somebody has to win the lottery! No. That's simply wrong. Or rather, it is true only for real lotteries. That's another of the shamefully wrong statistical arguments that darwinists love to use against ID. In lotteries, as one ticket is sorted out of all those which were sold, someone must necessarily win. But in a random search where th probability is extremely low, nobody will win even after billion of years. Think of that in this way: we have a lottery with 10^153 tickets. 10^3 tickets are sold. A winning number is extracted out of all the original 10^153 repertoire. The probability that one of those who bought a ticket may win are 10^-150, which is Dembski's UPB. That means that nobody would realistically ever win, even if one number were extracted each Planck time by each fundamental particle in the universe, for 14 billion years. So, no, it's not true that somebody has to win the lottery. You say: Organisms have the ability to evolve in *some* direction, or not at all. The probability of each direction might be very small but sometimes the organism *will* evolve and change over time in *some* direction even if the probabilty of each direction is small. No. Not if it is so small. Even considering the sum of all the probabilities for all directions. Finally, I can agree with you about the possibilities of dimerisation, or simply of exon shuffling, or of sexual allele shuffling: these "modular" reorganisations of existing domains have certainly an important role. I have never dealt with them, mainly because in those cases it is IMO much more difficult to compute the search space and the probabilities of a functional result. In principle, I can agree that at least some of these variations could be in the potential range of a random search. Others certainly are not. But anyway, they don't explain anything of how the basic information units (protein superfamilies) originated. That's why I stick to that model. I could vertainly discuss higher levels of organization (regulatory network, the immune system, the nervous system, or simply the genetic code) where the design is even more obvoious. But for many reasons, the rigorous treatment of those contexts is much more difficult. Therefore, I stick to signle proteins, where quantitative analyses are much more in our realistic range.gpuccio
September 10, 2010
September
09
Sep
10
10
2010
03:21 AM
3
03
21
AM
PDT
gpuccio, first of all, you are right about the citrate digestion. That makes this example a not so good illustration of IC! Well, in a way it still is. I think that most of the time parts of IC systems have or had other funtions than the IC one... Regarding Durston: I did not say that evolution can bring about *any* kind of function. My point is that evolution routinely finds *many* new ways to improve the fitness of organisms in different situations. Retrospectively looking only at the one that evolved and wondering about the small probability makes not much sense when the organisms could have evolved in completely different directions (or not at all). If you win the lotterey it also doesn´t make much sense to attribute this to divine intervention because the probability of this event is so small. Somebody has to win the lottery! Organisms have the ability to evolve in *some* direction, or not at all. The probability of each direction might be very small but sometimes the organism *will* evolve and change over time in *some* direction even if the probabilty of each direction is small. Also, it seems from your comment that you might accept that global information contents of organsims/genomes can increase. That is a good basis for future discussions about this topic I guess! Oh, and here is an example of how islands of functionality might be crossed. Since I am not a biologist the details are hard to understand for me, however! ;-)
Our study emphasizes how single point mutations can engender unexpected leaps in protein function thus enabling the appearance of new functionalities in proteins without the need for promiscuous intermediates.
http://nar.oxfordjournals.org/content/36/13/4390.full At least one other way to get new functionality without functional intermediates is by reactivating pseudogenes. This might be a rare situation however. On the other hand, intermediates might also be normal genes which undergo fitness-neutral changes until a new function emerges, again without intermediates with real function. Thirdly, sometimes a simple dimerisation can lead to new functions. Bovince seminal ribonuclease seems to be such a case, biologists might correct me if I am wrong.Indium
September 10, 2010
September
09
Sep
10
10
2010
12:51 AM
12
12
51
AM
PDT
They are certainly not darwinian mechanisms.
I'm not sure I follow. A discussion of artificial selection makes up a huge portion of Origin of Species. It's one of his main lines of evidence.Petrushka
September 9, 2010
September
09
Sep
9
09
2010
06:28 PM
6
06
28
PM
PDT
Petrushka (#117): You surprise me now. You should now quite well that I have no doubts about the huge possibilities of directed evolution and artificial selection. We have long discussed that issue, and you should remember that I am absolutely concinced that directed evolution and artificial selection are one of the best scenarios of intelligent design. They are certainly not darwinian mechanisms.gpuccio
September 9, 2010
September
09
Sep
9
09
2010
04:10 PM
4
04
10
PM
PDT
Directed evolution circumvents our profound ignorance of how a protein's sequence encodes its function by using iterative rounds of random mutation and artificial selection to discover new and useful proteins. Proteins can be tuned to adapt to new functions or environments by simple adaptive walks involving small numbers of mutations. Directed evolution studies have shown how rapidly some proteins can evolve under strong selection pressures and, because the entire 'fossil record' of evolutionary intermediates is available for detailed study, they have provided new insight into the relationship between sequence and function. Directed evolution has also shown how mutations that are functionally neutral can set the stage for further adaptation.
http://www.nature.com/nrm/journal/v10/n12/abs/nrm2805.htmlPetrushka
September 9, 2010
September
09
Sep
9
09
2010
11:51 AM
11
11
51
AM
PDT
gpuccio: Here's some more on the same subject: http://scholar.google.com/scholar?hl=en&lr=&q=related:snFAUWZhkIsJ:scholar.google.com/&um=1&ie=UTF-8&ei=WCuJTJy2MJSg8AT5-5TfDg&sa=X&oi=science_links&ct=sl-related&resnum=1&ved=0CCIQzwIwAAPetrushka
September 9, 2010
September
09
Sep
9
09
2010
11:48 AM
11
11
48
AM
PDT
Petruahka: I had already seen the link, but I need time to read the paper and understand it well. At first sight, it seems rather abstract and inconclusive, but give me time: if you have read the paper, you will have seen that it is rather complex. Anyway, I am happy that darwinists are going on in their efforts to falsify ID, while declaring that it is not a scientific theory. That's scientific debate...gpuccio
September 9, 2010
September
09
Sep
9
09
2010
11:33 AM
11
11
33
AM
PDT
gpuccio:
it appears possible for adaptive walks with only random substitutions to climb with relative ease up to the middle region of the fitness landscape from any primordial or random sequence
See the link at post #105Petrushka
September 9, 2010
September
09
Sep
9
09
2010
10:42 AM
10
10
42
AM
PDT
Indium (#111): I do want.gpuccio
September 9, 2010
September
09
Sep
9
09
2010
08:39 AM
8
08
39
AM
PDT
Indium: Thank you for your comments. I respect your opinions, I will just explain briefly why I don't think they are pertinent: 1) Your only explicit objection to Durston's metric is the old (and IMO completely inconsistent) concept that evolution can produce "any kind of function". That's false. As I have argued many times, complex systems cannot do with "any kind of function". They pose severe restrictions to what can be useful and what cannot. Protein folds and active sites must be very specific give those biochemical properties which can be integrated into the complex network which already exists, not to speak of regulatory networks and procedures, and so on. So, in each defined context, evolution can really take only a few specific directions, even for a conscious engineer, even more for mere chance and NS. And anyway, the vast majority of protein sequences remain non functional in any given biological context. 2) You can believe what you like. That's not the same as showing that it is a credible scientific hypothesis. Anyway, I never question people's faith. 3) "In principle", stranger things are possible. ID is not about what is impossible "in principle" (ID theory is not a mathematical deduction). ID, and empirical science, is instead, about what is "empirically" impossible (or possible, or likely). If you are not interested in empirical science, it's your option. 4) Look, you say that because of general probability estimations evolution can *in principle* not add, say 150 bits of information. I never, never said that. I say that a function which requires more than 150 new bits of functional information to appear from a starting state cannot be empirically found through pure random variation. I don't believe it's the same thing. This is easily refuted: When a process can add 10 bits, by repetition it can add 150 bits or 1000 bits. You are easily refuting what you had easily imagined I had said. My compliments. 5) Finally, Lenski again. You say: Let´s go back to the Lenski case: The new citrate permeability is only usefull because these organisms can already digest citrate, which in itself is a complex function, right? How many bits do we need for citrate digestion? 10? 20? 50? So, how is the evolution of citrate permeabilty unrelated to the previous evolution of citrate digestion? Why shouldn´t we add the 10 bits to the previous information to get the information content for the full system? What happens if a subsequent evolutionary step makes these processes 100 times more efficient? Another 10 bits? And so on. Let´s imagine Lenski would not have done this work. I am sure at some point some ID guy would have come along, seen this system and decided that this is a (small) unevolvable IC system: How usefull is citrate digestion without permeability and vice versa? I think you are apparently confounded here. Living beings use citrate all the time. It is the essential component of the Krebs cycle, which is universal in all aerobic living cells. Again I quote Behe: "Now, wild E. coli already has a number of enzymes that normally use citrate and can digest it (it’s not some exotic chemical the bacterium has never seen before). However, the wild bacterium lacks an enzyme called a “citrate permease” which can transport citrate from outside the cell through the cell’s membrane into its interior. So all the bacterium needed to do to use citrate was to find a way to get it into the cell. The rest of the machinery for its metabolism was already there. As Lenski put it, “The only known barrier to aerobic growth on citrate is its inability to transport citrate under oxic conditions." So, your question "How usefull is citrate digestion without permeability and vice versa?" is easily answered: very useful indeed! The problem is only that E. coli cannot use exogenous citrate, for lack of permeability to it. Your comments about possible wrong inferences of IDists about IC are therefore completely out of order. (And anyway, I have never used the concept of IC in my discussions with you, because it was not necessary for my argument here). Moreover, I thought I had made it clear that nowhere in my argument I was measuring the global information content of a whole organism. So, why do you continue to argue in that sense? I have discussed the difficulties in the evolution of protein domains by darwinian mechanisms, applying the concept of CSI to single proteins, and never to more complex systems. Therefore, I consider your objections about the total information content of an organism absolutely irrelevant.gpuccio
September 9, 2010
September
09
Sep
9
09
2010
08:37 AM
8
08
37
AM
PDT
Oh, btw, if you want I can dig up a paper where it is shown that single point mutations can indeed generate a large jump in binding properties, which I think would invalidate all your comments about uncrossable islands of functionality.Indium
September 9, 2010
September
09
Sep
9
09
2010
08:24 AM
8
08
24
AM
PDT
gpuccio, thanks again for your explanation. I think I now have to divide my answer into some kind of bullet points: 1) No, I don´t accept Durstons metric at all. We can go into a detailed discussion of this if you like, but you will also find effective rebuttals on the web. One of the basic problems is that he is subject to some kind of lottery fallacy: He just checks one potential outcome or function. Evolution could have resulted in an incredibly large set of different functions and also in completely different realizations of the same function and each time Durston retrospectively would calculate the amazingly low probability that exactly THIS outcome is observed. 2) I believe that very simple mutation events can open roads to completly new and unrelated protein-protein binding sites and therefore to new functions. Also, large unselectable areas of sequence space can still be crossed by neutral steps or pseudogenes which are later reactivated. 3) If evolution can add 10 bits of information to the genome of an organism, it can also add 150 bits or 1000 bits. It just takes time. Nothinm that you say changes this fact *in principle*. Since the general principle is all I am interested in at the moment, this is enough for me. Look, you say that because of general probability estimations evolution can *in principle* not add, say 150 bits of information. This is easily refuted: When a process can add 10 bits, by repetition it can add 150 bits or 1000 bits. If you now say that the organisms somehow might loose information that is a completely different argument! Let´s go back to the Lenski case: The new citrate permeability is only usefull because these organisms can already digest citrate, which in itself is a complex function, right? How many bits do we need for citrate digestion? 10? 20? 50? So, how is the evolution of citrate permeabilty unrelated to the previous evolution of citrate digestion? Why shouldn´t we add the 10 bits to the previous information to get the information content for the full system? What happens if a subsequent evolutionary step makes these processes 100 times more efficient? Another 10 bits? And so on. Let´s imagine Lenski would not have done this work. I am sure at some point some ID guy would have come along, seen this system and decided that this is a (small) unevolvable IC system: How usefull is citrate digestion without permeability and vice versa?Indium
September 9, 2010
September
09
Sep
9
09
2010
07:12 AM
7
07
12
AM
PDT
zeroseven: I think that in this thread (and maybe also in others) I have answered in some detail, and I hope with some clarity, to many of your questions (for instance, in 85 and 88). You have given not one word of comment about my answers. But, fortunately, you have now creatively managed a new question about chihuahuas. Is that your usual epistemological approach to discussions? I hope you are enjoying yourself, my friend... :)gpuccio
September 9, 2010
September
09
Sep
9
09
2010
04:15 AM
4
04
15
AM
PDT
Indium: First of all, the Szostak paper. I have given a very detailed analysis of it and of its cerdibility and meaning on this blog. I have also debated my analysis on the same thread with one very competent and thoughtful interlocutor who has done his best to defend the paper (almost certainly a very good biologist). At the end of the discussion, I am still absolutely convinced that the paper is seriously flawed in its conclusions, and that I have given very good evidence of that. In this moment I don't remember what thread it was (maybe someone can provide a link, or I will have to look for it later and give you the link). You can read the whole discussion and judge for yourself. Let's go to your questions: What stops an organism from adding this kind of information (like in Lenskis lab) over time? What stops these evolution events from being dependent on each other or from being related, for example in a way that the combined change leads to a completely new effect? These are two different questions. My answer to the first is: nothing stops an organism from undergoing microevolutionary events which are compatible with its probabilistic resources for the random part, and which can be selected in the context of its environment. That happens, even if it usually requires, at least in the observed cases, a very high reproductive rate, big populations, very small complexity of the transition (one or two AAs), and a very strong envronmental selection (think of antibiotic resistance) to happen in an observable time (which, however, can be of many years, especially for supposed two AAs mutations: selectable one AA mutations can be achieved in a bacterial culture in short times). You can find a very good discussion about those empirically observed facts in Behe's TEOE. What you can have, in the end, is a certain number of microevolutionary events, bearing small unrelated tweakings of existing functions, in the organism. To the second question, my answer is: it is not a question of being "stopped": functions either are related, ore are not. Let's be more clear: At a certain time in natural history, a new protein domain representing a completely new protein superfamily appears, in a new species. That has happened thousands of times, according to our knowledge of the proteome, and of natural history. Let's call this new protein domain A, and let's assume that its length is about 130 AAs (a very reasonable length for a protein domain). To be more precise, let's assume that we can apply the Durston method to compute the real functional information in that domain: let's say it is 350 Fits (a reasonable value: in Durston's paper, protein families of that length have approximately that functional information). So, it is perfectly legit to ask: how did that new protein domain arise? The darwinist answer will probably be that it was the result of gradual mutations, possibly in a duplicated gene of some pre-existing protein domain. Let's call that "precursor" A. So, A exists before B, and B derives from A, in our model. For simplicity, we can think of A and B as approximately of the same length. But, as they are domains in two different protein superfamilies, by definition they are unrelated at the primary sequence level: we can safely assume that they present less than 10% homology, which is the same as a completely random level of homology. IOWs, A is a sequence completely isolated from the sequence of B in the search space, and is in no way "near" B. These are all things we know. So, the transition from A to B, if it is completely random, must create 350 bits of functional information by a random walk in the search space: that is empirically impossible. You say: but a series if related small functional variations could bring us from A to B. I ask you: why should it be the case? You should give me at least one of two kinds of arguments to make me believe that such an assumption is credible, at least as hypothesis: 1) Give me some logical reason why it should be the case: I cant' see any. We have two long sequences, totally unrelated and isolated in the search space, with two completely different folds and functions. What reason in the world can you suggest for them being connected by a series of small functional states, each one selectable? There is nothing in what we know of protein folding, of protein function, and of biochemical laws, which can justify that. And that strange property should be true not only in one case, but in thousands of unrelated cases. 2) You could give me empirical evidence: you could say: look, I have this example, or at lest this model, where I have detailed the 50 or 70 intermediate functional states and shown why tey are selectable, and how thet realize a series which "builds" the new sequence for the new final function. Can you, or can anyone else? Or, to put it differently: Based on the ability to add small amounts of information, can we agree that the information content of an organism is not constant and can be increased through evolution? You have really said something different here. There is no doubt that the information content if an organism is not constant. Each new genetic disease is due to the loss of some functional information. In rare cases, and in specific contexts, microevolutionary events create a few bits of new functional information. We agree on that. What we probably don't agree about, is that the complex individual functions that we observe today in the proteome could have originated sequencially from existing precursors by a darwinian mechanism. I deny that model for two important reasons: a) Each single transition to generate a new protein domain/superfamily is too complex to be in the range of purely random variation. b) There is absolutely no rational motive and no empirical evidence that complex functions, such as protein domains, can be achieve through a sequence of small functional and selectable variations. Indeed, all we know and observe is absolutely against that.gpuccio
September 9, 2010
September
09
Sep
9
09
2010
04:09 AM
4
04
09
AM
PDT
Ok it doesn´t! ;-) http://www.nature.com/nature/journal/v410/n6829/full/410715a0.htmlIndium
September 8, 2010
September
09
Sep
8
08
2010
11:29 PM
11
11
29
PM
PDT
gpuccio, thanks for the explanation. You certainly have a point: Calculations of the probabilities are difficult and it is probably not always warranted to just add the different bits. Still, in a way that doesn´t safe your argument. If, repeatedly, new functions with small amounts of information can arise in an organism it will over time become more complex (and contain more information, right?). And since there is no reason why subsequent evolutionary changes might not be dependent on previous ones, this will in retrospective give the impression of irreducible complexity or at least of a very improbable event. In a way, that is exactly what seemed to have happened in Lenskis lab. I have to ask again: What stops an organism from adding this kind of information (like in Lenskis lab) over time? What stops these evolution events from being dependent on each other or from being related, for example in a way that the combined change leads to a completely new effect? Or, to put it differently: Based on the ability to add small amounts of information, can we agree that the information content of an organism is not constant and can be increased through evolution? Do you know ? (Hope the link works!)Indium
September 8, 2010
September
09
Sep
8
08
2010
11:28 PM
11
11
28
PM
PDT
The fitness landscape in sequence space determines the process of biomolecular evolution. To plot the fitness landscape of protein function, we carried out in vitro molecular evolution beginning with a defective fd phage carrying a random polypeptide of 139 amino acids in place of the g3p minor coat protein D2 domain, which is essential for phage infection. After 20 cycles of random substitution at sites 12–130 of the initial random polypeptide and selection for infectivity, the selected phage showed a 1.7×104-fold increase in infectivity, defined as the number of infected cells per ml of phage suspension. Fitness was defined as the logarithm of infectivity, and we analyzed (1) the dependence of stationary fitness on library size, which increased gradually, and (2) the time course of changes in fitness in transitional phases, based on an original theory regarding the evolutionary dynamics in Kauffman's n-k fitness landscape model. In the landscape model, single mutations at single sites among n sites affect the contribution of k other sites to fitness. Based on the results of these analyses, k was estimated to be 18–24. According to the estimated parameters, the landscape was plotted as a smooth surface up to a relative fitness of 0.4 of the global peak, whereas the landscape had a highly rugged surface with many local peaks above this relative fitness value. Based on the landscapes of these two different surfaces, it appears possible for adaptive walks with only random substitutions to climb with relative ease up to the middle region of the fitness landscape from any primordial or random sequence, whereas an enormous range of sequence diversity is required to climb further up the rugged surface above the middle region.
http://www.plosone.org/article/info:doi/10.1371/journal.pone.0000096Petrushka
September 8, 2010
September
09
Sep
8
08
2010
04:20 PM
4
04
20
PM
PDT
zeroseven you ask: Does a chihuahua have a different amount of CSI than a wolf? Yes: Phylogenetic Relationships, Evolution, and Genetic Diversity of the Domestic Dog Excerpt: The Xoloitzculntli or Mexican hairless dog also has gone through population contraction followed, presumably, by close inbreeding for several hundred generations. Thus it is likely to have reduced genetic variation,,,,,,the mean sequence divergence in dogs, 2.06, was almost identical to the 2.10 (sequence divergence) found within wolves. (please note the sequence divergence is slightly smaller for the entire spectrum of dogs than for 'parent' wolves) http://jhered.oxfordjournals.org/cgi/reprint/90/1/71.pdf But this is not surprising zeroseven in that we find that Natural Selection reduces genetic diversity because it 'sifts' what is already preexisting genetic information without ever replenishing the loss of genetic diversity: "...but Natural Selection reduces genetic information and we know this from all the Genetic Population studies that we have..." Maciej Marian Giertych - Population Geneticist - member of the European Parliament - EXPELLED Another strong piece of genetic evidence, for the recent origin of man, is that scientists find the differences of the 'younger' human races (Chinese, Europeans, American Indians, etc.. etc..) are losing genetic information when compared to the original race of humans which is thought to have migrated out of east Africa some 50,000 years ago. "We found an enormous amount of diversity within and between the African populations, and we found much less diversity in non-African populations," Tishkoff told attendees today (Jan. 22) at the annual meeting of the American Association for the Advancement of Science in Anaheim. "Only a small subset of the diversity in Africa is found in Europe and the Middle East, and an even narrower set is found in American Indians." Tishkoff; Andrew Clark, Penn State; Kenneth Kidd, Yale University; Giovanni Destro-Bisol, University "La Sapienza," Rome, and Himla Soodyall and Trefor Jenkins, WITS University, South Africa, looked at three locations on DNA samples from 13 to 18 populations in Africa and 30 to 45 populations in the remainder of the world.- I wonder what Hitler would have thought of that study? This following study is interesting in that it shows the principle of Genetic Entropy being obeyed for the estimated 60,000 year old anatomically modern humans found in Australia: Ancient DNA and the origin of modern humans: John H. Relethford Excerpt: Adcock et al. clearly demonstrate the actual extinction of an ancient mtDNA lineage belonging to an anatomically modern human, because this lineage is not found in living Australians. Although the fossil evidence provides evidence of the continuity of modern humans over the past 60,000 years,,, http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=33358 EXPELLED - Natural Selection And Genetic Mutations - video http://www.metacafe.com/watch/4036840 Natural Selection Reduces Genetic Information - Dr. Georgia Purdom - video http://www.metacafe.com/watch/4036808 Natural Selection Reduces Genetic Information - No Beneficial Mutations - Spetner - Denton - video http://www.metacafe.com/watch/4036816 Darwinism’s Last Stand? - Jonathan Wells Excerpt: Despite the hype from Darwin’s followers, the evidence for his theory is underwhelming, at best. Natural selection - like artificial selection - can produce minor changes within existing species. But in the 150 years since the publication of Darwin’s Origin of Species by Means of Natural Selection, no one has ever observed the origin of a new species by natural selection - much less the origin of new organs and body plans. http://www.evolutionnews.org/2009/06/junk_dna_darwinisms_last_stand.html#more further note; This following paper, though of evolutionary bent, offers a classic example of the effects of Genetic Entropy over deep time of 270 million years: A Cambrian Peak in Morphological Variation Within Trilobite Species; Webster Excerpt: The distribution of polymorphic traits in cladistic character-taxon matrices reveals that the frequency and extent of morphological variation in 982 trilobite species are greatest early in the evolution of the group: Stratigraphically old and/or phylogenetically basal taxa are significantly more variable than younger and/or more derived taxa. http://www.sciencemag.org/cgi/content/abstract/317/5837/499bornagain77
September 8, 2010
September
09
Sep
8
08
2010
03:48 PM
3
03
48
PM
PDT
Indium: maybe I have not been clear enough. In Lenski's work, always according to our assumptions, a couple of mutations changed some existing protein a little, allowing permeability to citrate. Microevolutionary changes usually don't change substantially the protein, they just "teak" an existing protein in its island of functionality. Again, I don't know what the molecular basis of Lenski's function is. So all this discussion is higly speculative, but it can be valid as a general model. So, let's say that protein A changes a little because of a coordinated mutation of two aminoacids, which bears a new function form its existing basic structure and fold (usually, the easiest way for that to happen is a change at the active site). Well, we have a transition which has given a new function. So, the transition is functional, specified. Is it an example of dFSCI? No, because the complexity of the transition is, at most, of 8.65 bits (it could be less). As you say, the mutation, at least in Lenski's case, has been fixed. That's because it gives a reproductive advantage (at least in Lenski's environment). As I have already mentioned, that process requires two events: 1) The original bacterium where the double mutation occurred must expand to all, or most of the population, thanks to its reproductive advantage. 2) In that process, the mutations must be preserved by negative selection of new mutations on those sites. Point 1 is certainly the most important in a short term scenario, but both are anyway in the range of NS, because the double mutation has given a true advantage to its carrier, at least in that context. I think we can agree on that. That's what is called a microevolutionary event. Now, you say: After that fixation nothing prevents these new strains from undergoing another round of the same mutation and fixation cycle, correct? Yes, nothing prevents the new population, with its citrate permeability, to undergo some other microevolutionary event of one or two mutations in the same individual, bearing some other new tweaked function. And so? And then they would have increased the information content by 20 bits. No, that is a common mistake. A transition of 20 bits is a transition where 20 new bits of functional information are necessary to achieve a new function from the initial state. It is a transition wherte at least 5 coordinated mutations must occur randomly in the same individual, to bring the functional variation which can be selected. That is not the sum of two independent, and unrelated, microevolutionary events of 10 bits each. The probability of each 10 bits event is 1:1024. But the probability of a single 20 bits event is 1:1048576. It's 1000 times more unlikely. Functional bits are an exponential measure. So, if 20 bits are necessary to achieve a new function, that event will occur in about 1000 times the time in which a single 10 bits event can occur, if all the other variables (population size, mutation rate, etc.) remain the same. That's why my threshole for single functional random variations is of 150 bits (about 35 AAs necessary for the function). That is a very good threshold to make a random event completely unlikely in a conceivable biological context. Whilw I have lowered the value from Dembski's original 500 bits, that is still an extreme threshold in our context. So, how could a 4 AAs variation (about 17-18 bits) be achieved in a reasonable time? There is only one possibility. Let's say that A must become B to give a new function. And let's say that the difference is of 4 AAs. The times implied would be really long. But if a state exists, let's call it A1, which is intermediate between A and B and is different form them for only two AAS, and that state is functional and selectable, then the transition A - B can be deconstructed into two independent transitions: A - A1 and A1 - B, each of two AAs. A1, when achieved in one individual of the population, would expand, and become a new population where the second part of the variation could happen with the same probability as the first: IOWs, the total probability of the event would be much higher than in the case where all four mutations must be achieved randomly in the same individual for a new function to arise. Is that clear? That's why I have written, many times, that: "more complex mutations must be deconstructed into simple selectable steps for the darwinian model to work". Now, you can believe that such a deconstruction can be done in all cases. I don't. And there are many reasons not to believe that. The first is that there is no logical reason to assume that. We know that complex functions are not the passive sum of simpler functions, but require higher levels of organization tro work. that is generally true in all fields. In the biological field, in particular, there are a lot of reason to understand that the darwinian myth of function deconstruction is, indeed, a myth. Let's go back to our example. There are different ways in which our A1 can be selectable and expandable. And please, remember that, to be expanded, A1 has to be more functional than A, because it's exactly A that it must expand against. So, the possibilities: 1) A1 can have the same function of A, but at a higher grade. This is simplre enough, and easiest "tweak" which can be realized. If B too has the same function at a higher level than A1, the whole transition is simple enough. But the problem is that it does not create a real new function: it just improves what already exists. To be more clear, as we are speaking of proteins, that case is a case of moving inside a functional island, and a specific fold, at most improving the affinity of an existing molecule for its substrate, or shifting to a similar substrate of the same kind, but slightly different. Note that most of the few examples we have of really understood microevolution are of this kind (and nylonase would be one of them). 2) A1 can have a completely different function, and fold. I don't believe that any of that can be achieved with a two AAs mutation, certainly not starting from a totally unrelated different protein superfamily. But if it could, two possibilities remain: 2a) the transition form A1 to B remains in the same new functional island. then, the really important transition is only the foirst. the second, again, is only a tweak of what already exists. 2b) The second transition again finds a new functional configuration and fold with only two more AAs change. Woderful. What a pity that it does not happen, neither for the first, nor for the second transition. What really has happened, and we don't know how (at least according to the darwinian model) is that new protein domains and superfamilies have constantly appeared in natural history, that each of them is totally unrelated to the others, and that the transtition from one to the other would imply a jump in functional information well above my threshold of 150 bits. And no functional intermediates are known, in any case of that kind of transitions: they are not in the proteome, they have not been proposed by darwinists at any detail of molecular model. They exist only in their imagination. A myth, as I said. In the emantime, darwinists play with fairy tales of small steps which nobody has ever really conceived, least of all observed. Of neutral mutations which become fixed and hey, they are magically just those we needed to "help" the tired mechanism of NS. Of proteins which traverse the almost infinite search space of protein sequences as though they could proudly swim in it and gain the distant shore. And so on. While the facts are: more than 6000 protein domains totally unrelated at the sequence level (less than 10% homology), each of them functional and represented by multiple functional proteins in the proteome, and no intermediate newteen them ever observed. 35 protein families analyzed by Durston for their functional information content, which is in the range of 46 (ankyrin) to 2416 (Flu PB2) Fits, with 28 of them above my threshold of 150 Fits. And so on.gpuccio
September 8, 2010
September
09
Sep
8
08
2010
03:03 PM
3
03
03
PM
PDT
gpuccio, Does a chihuahua have a different amount of CSI than a wolf?zeroseven
September 8, 2010
September
09
Sep
8
08
2010
01:57 PM
1
01
57
PM
PDT
gpuccio, in Lenskis work the change you quantified to have roughly 10 bits was fixed in populations of bacteria. After that fixation nothing prevents these new strains from undergoing another round of the same mutation and fixation cycle, correct? And then they would have increased the information content by 20 bits. Or is there anything that prevents the new strains from evolving after the first 10 bits? A hidden switch?Indium
September 8, 2010
September
09
Sep
8
08
2010
11:28 AM
11
11
28
AM
PDT
1 2 3 4 6

Leave a Reply