Ken Miller, the honest Darwinist

Patrick: "I think someone needs to summarize the argument and conclusions in this thread." I have tried to summarize my take on this discussion with a further elaboration of the numerical argument: Nicholas White in his 2004 paper makes an estimate of the rate of appearance of chloroquine resistance in malaria Plasmodia. "Resistance to chloroquine in P. falciparum has arisen spontaneously less than 10 times in the last 50 years (14). This suggests that the per parasite probability of developing resistance de novo is on the order of 1 in 10^20 parasite multiplications." This is cited by Behe in TEOE. An estimate for the actual parasite multiplication rate per 48 hours (PMR) is given as about 8. With the exponential growth curve and no die off this leads to a 1 trillion population in about 28 days. There are in any year about 42-43 million people sick with malaria. If it takes about 5 years worldwide to develop resistance somewhere, this is about 70 periods of multiplication to a 10^12 population in 4.25 x 10^7 people. This is 1 in a total of 7.0 x 10^13 x 4.25 x 10^7 = 2.3 x 10^21 parasite multiplications. The initial stages of parasite multiplication may be against the full uncompromised immune system attack, where White's 2-3% figure applies. This will stretch out the time required for the clone to reach "transmissable densities". During this preliminary period of the infection the immune system will destroy some percentage of spontaneously occurring CQR mutated parasites before they can multiply by many times. So this is an attrition factor that stretches out the time to achieve the 1 trillion population. To be conservative, we could assume it increases the time period to a 10^12 population from 28 days to 280 days. This would reduce the number of multiplications to 2.3 x 10^20. But this number still does not reflect the CQR parasites that were killed by the immune system. We can assume that the immune system was not significantly weakened up to the point of transmissible densities. This would be up to say a population of 500 million. During this period say 98% die. But during this stage the immune response factor is insignificant because it takes place in relatively small populations - the vast bulk of the multiplications take place after this point, in the acute phase leading from 500 million or so to 1 trillion. In this latter part of the progression of the disease the population expands by a factor of about 2000. In the latter stages of the infection leading to the 1 trillion population the host's immune system is compromised, so during this stage which includes most of the total multiplications the immune system kill off is much less than 98-97%. There is one more factor: partial reversal of fitness selection due to the CQR parasite being some percentage less fit than the wild variety in a non drug environment. This also would reduce the chances of a CQR mutant surviving to propagate into a large population. This factor is also insignificant because the people being considered in the estimate are all being treated with chloroquine, in order to clinically observe chloroquine resistance. As a result, in this human population sick with malaria during the time being considered, the CQR parasites are always favored over the normal type. This reasoning attempts to figure in all the objections, but it still results in approximately 1 in 10^20 to 10^21 multiplications. magnan

I should have said "...2-3% of the parasites in each parasite life cycle (not each multiplication) to generate a variation of the cell surface antigen" magnan

PaV " I took that for a maximum number. But, of course, if you multiply 5 years (avg. time for resistance to develop) and 183 (365 divided by the two-days needed to develop the 10^12 parasites in each of the severely ill malarial patients), this leads to 0.915 x 10^20 malarial parasites needed for CQR to develop." It is good to try to come up with a more rigorous estimate. I would differ from this a little. The parasite life cycle of invading a red blood cell, multiplying inside and then bursting to spread more parasites is about two days. I found an estimate for the actual parasite multiplication rate per 48 hours (PMR). This is given as about 8. With the exponential growth curve and no die off at all this leads to a 1 trillion population in about 28 days. If it takes about 5 years worldwide to develop resistance somewhere, this is about 70 periods of multiplication to a 10^12 population in 4.25 x 10^7 people. This is 1 in a total of 7.0 x 10^12 x 4.25 x 10^7 = 2.3 x 10^21 multiplications. magnan

Jehu: "You say White put the odds of selection at 2-3%? If White really believed that he should have estimated the number of de novo CQR events at 250 instead of “less than 10.” I think the odds of selection in a sick person is about 50%, which fits with White’s calculation." This is a good point. But he is clear on the immune response factor. "It is likely that this specific immune response directed against the immunodominant surface antigens will reduce the probability of the usually single mutant parasite ever multiplying sufficiently to transmit as for P. falciparum; there is only a 2-3% chance that the genetic event causing resistance would arise in the antigenically variant subpopulation that will expand to reach transmissable densities." The parasite apparently has a systematic way of evading the host immune response. This is by programming 2-3% of the parasites in each multiplication to generate a variation of the cell surface antigen that the immune system responds to. Very clever. Perhaps White didn't figure this in to his calculation because he was mainly considering acute phase infections, after the parasite has already reached "transmissible" densities (apparently that is in the order of hundreds of millions to a few billion Plasmodia, not the trillion or so for acute phase. During the latter period presumably the host immune system is weakened considerably, so the 2-3% survival rate no longer applies. magnan

gpuccio: "....any contribution of other factors (immune system of the host, reversal of mutation for fitnell loss, etc.) cannot explain the great difference in orders of magnitude between observed CQR and other observed resistances." The host immune response and (partial) fitness reversal effects can be considered to in principle commonly apply to both single and double mutation resistance genetic events. So the observed huge difference of frequencies of occurrence would not be expected to be affected. gpuccio: " But, even if one of the alternative views were partially true, Behe’s thesis is more than valid." I agree. I just think it is prudent to carefully examine all possible criticisms. As I mentioned, I think this particular objection is actually technically valid. It directly follows from the way White apparently did his calculation of the frequency of "de novo" resistance genetic events. But it is a minor point that still doesn't affect Behe's thesis. gpuccio: "It could, alternatively, have developed a metabolic way of destroying the drug “chloroquinase”, or something like that)......Or the plasmodium could have developed a new pump system, out of the blue, with the only function of getting rid of the drug." The only way critics could deal with this would be to show that there are no more elaborate strategems possible to the parasite, presumably due to its basic structural design. In other words, that your suggestions are impossible without fundamentally redesigning the whole organism. This is ridiculous on the face of it. magnan

Is there some resistance in other animals to malaria that is no present in humans? Could it be that we "thin-skinned" humans are easy targets for the probing mosquito proboscis? PaV

I think someone needs to summarize the argument and conclusions in this thread and post it as a front page thread. Patrick

You might be confusing the total number in a year with the total number on any given day. White gives the number of between 10^16 and 10^18 as the population in sick people in any two-day period, not over the course of a year.---------Jehu I was working off of Figure 5 from White's "Antimalarial Drug Resistance". It shows 10^17 as the number of parasites "in the world today". I took that for a maximum number. But, of course, if you multiply 5 years (avg. time for resistance to develop) and 183 (365 divided by the two-days needed to develop the 10^12 parasites in each of the severely ill malarial patients), this leads to 0.915 x 10^20 malarial parasites needed for CQR to develop. Seems awful close to 1 in 10^20, doesn't it? :) So I'm much more comfortable with the numbers. And, alas, for the Darwinists: they've got to come to terms with these numbers. PaV

I have a question. What is unique about human blood that makes the malaria parasite go after it and not other animal's blood? Is there some resistance in other animals to malaria that is no present in humans? jerry

jehu and magnan: thank you for your very thoughtful attempts at verifying thsoe calculations. I appreciate them very much, and they have certainly contributed to clarify some points. I think that the important thing is that Behe's thesis is based essentially on a fact: the rarity of chloroquine resistance compared to other malaria resistances. The best explanation for that is that it is a more complex mutation, requiring "at least" two independent, selectable mutations. That explanation, and White's calculations, all fit very well the data. The objections of JAM (I am sorry that he is no more here, even if I agree that his attitude was very destructive, to the point of disrespect for others) and of other darwinist are, it seems, of two kinds: 1) White's (and Behe's) statement that CQR happens so rarely (1 in 10^20 individuals) is wrong. Maybe CQR happens much more frequently. I think this objection is unsubstantiated and doesn't fit the data. CQR could happen more frequently if it were a single mutation, which it is not, or if it were the result of two independent, sequential, selectable mutations, which is not in accord with the data. JAM, I think, tried to support both these views (which is contradictory). The first point is simply not true, at least if we refer to clinically significant resistance, whcih is the only one which, in the real world, has produce a true fixed mutant. In all isolated CQR forms, if I understand well, we find at least two mutations, sometimes more. Besides, if CQR were the product, at least in some cases, of a single mutation, that mutation should absolutely be found much more frequently than all others, and I really believe that any contribution of other factors (immune system of the host, reversal of mutation for fitnell loss, etc.) cannot explain the great difference in orders of magnitude between observed CQR and other observed resistances. 2) But, even if one of the alternative views were partially true, Behe's thesis is more than valid. The fact remains that, in the real world, an organism (malaria plasmodium) in the presence of a very strong fitness landscape constraint (chloroquine generalized use, or if you want S hemoglobin), and with a huge probability resource available (a very high number of replications) has developed nothing more than "burning the bridge" adaptations, all of them molecularly very simple (one or a few more "destructive" mutations), none of them building true new information. That's a fact, and it is not really important if White's calculations be one or two orders of magnitude wrong. The malaria plasmode, in all that time, with all those replications, and using all possible tools of randomness and selection, including both asexual and sexual reproduction, has manages just that. Darwinian evolution has managed just that. Now, maybe malaria plasmodium, or HIV, or E. coli, are not loved by darwinian evolution, and so they are purposely kept out of its best resources. The fact is that humans, with infinitely less resources, are supposed to have developed in a short time all the adaptations which explain the difference with chimps, however big you think they are. I just want to answer here to another objection that I have read, that the plasmodium could only adapt to chloroquine with those "simple" mutations, while humans could have benefited of much more complex mutaations for their evolutionary goals. That's not true. If plasmodium evolved that way it was not because it was the only way, it was the only "simple" way. It coul, alternatively, have developed a metabolic way of destroying the drug (a "chloroquinase", or something like that). Bacteria have penicillinases, but they have not evolved by random mutation. They are simply exchanged by horizontal gene transfer. Or the plasmodium could have developed a new pump system, out of the blue, with the only function of getting rid of the drug. Or it could have adapted its own metabolism (not only the drug target) so that the drug could no more affect it, for instance developing a new pathway to get rid of the heme molecule. Or it could have become a completely different organism, a new species, more efficient as a parasite, to avoid the S hemoglobin menace. Or it could have become a multicellular organism, or an independent, non parasitic organism, or just a parasite of something else. Or... Just mention what you want. After all, with lower evolutional resources, chimps are supposed to have become men. And many other magic adaptations are supposed to have taken place. And, if CQR can happen with only one mutation, which is not true, Behe's thesis stays valid: why only so simple mutations have taken place? Why nothing more complex, more elegant, more efficient? Why is the blind watchmaker so blind that he cannot produce any watch? And if the CQR mutation happens more frequently, the same questions apply. Why so poor results? If mutation reversal is so powerful a force, isn't that a further proof of how darwinian mutations are almost always negative, or neutral at best, even when they bring some advantage? In the end, the only truth is this: Behe is right, completely right. gpuccio

Magnan, The number White gives is "less than 10" for de novo CQR events. As of 2002 only 4 de novo CQR events were known. It appears that another was discoverd in the Philipines. That makes 5. So "less than 10" is twice as many as were actually discovered. You say White put the odds of selection at 2-3%? If White really believed that he should have estimated the number of de novo CQR events at 250 instead of "less than 10." I think the odds of selection in a sick person is about 50%, which fits with White's calculation. Jehu

In his 2004 paper White estimates the effect of specific immune response. It boils down to there being a 2-3% chance that the mutant parasite will evade the host's specific immune response and expand in population to "transmissable densities". So my guess of 1% is very conservative at 1/2 - 1/3 the expert estimate of the factor, making it unlikely to be relevant. magnan

Jehu: "You really have to allow very slim chances of selection to dent those numbers. Since we are only talking about sick people, that criticism doesn’t really hold water." The point is that this estimate is technically for the number of CQR genetic events in parasites that survived past host immune response and differential fitness selection. I just wanted to verify that this factor really was irrelevant. I think you are right, but I just wanted to look at some numbers. For this issue the crucial parameter in the calculation is the number of occurrences of CQR reported worldwide. If there were no such selection factors killing off some occurrences of CQR in individual parasites during the period leading up to actual clinical detection of patient CQR, how many reports of CQR in patients would there have been? With 10 reports actually and if the "die-off" factor is 99 out of a hundred (1% survive), with no die off there would have been 10 x 10^2 = 10^3 = 1000 reports. Then instead of 1 in 10^20 - 10^21 the estimate would be 1 in 10^18 - 10^19. Still very low and leaving Behe's argument essentially untouched. This "CQR parasite die off factor" must be limited (and is probably less than the 99% assumed in the above calculation), because during the sickness the immune system is compromised so immune system attack is not a large factor killing off CQR parasites, and during much of the course of the disease chloroquine is being administered, so selection is definitely for the CQR strain and not being periodically reversed to favor the non-CQR strains. So it seems to me this is a technical objection that is not germain to the validity of Behe's thesis. Any critic would have to show these effects to be many orders of magnitude greater to make the objection credible. magnan

PaV

The calculation I made is from memory and comes from numbers in another paper by White. The 10^12 number of parasites is true ONLY for those who become seriously ill with malaria—and probably end up dying. That represents just a percentage of all those who are infected. That’s why for (b) I used the figure of 4 to 5% of the world-wide number infected.

Let's see if we can tighten the numbers up a bit. The number of sick people each year is about 4.25 x 10^8. (WHO report) The median number of parasites in a symptomatic person at any given time is 2.5 x 10^11 (Dondorp 2005) Over 50 years CQR has occured de novo less than 10 times. So our calculation is [(4.25*5*2.5)*(10^8*10^11*10^1)]/10. So that gives us a per parasite probability of 1 in 5.31 x 10^20. That is over five times lower than the probability White and Behe give. However, I think the actual probability is even lower than that . The number I used for the parasite population of sick people is the biomass on any given day. The actual number in question is the number of reproductive events over the term of the sickness. Remember, we are looking for opportunities to mutate to CQR. I invite you to challenge my math. I'm no William Dembski you know.

As I also pointed out, though, this number becomes around 10^19 parasites in the whole world in any one calendar year—-which is in contradiction of White’s number of 10^17 for the total number of parasites. So, there’s a discrepancy. And, Behe’s number could possibly be criticized for being too high.

You might be confusing the total number in a year with the total number on any given day. White gives the number of between 10^16 and 10^18 as the population in sick people in any two-day period, not over the course of a year. Jehu

Looks like I made an error, the population of sick people is 4.25 x 10^8. So, according to my calculations, that would put the de novo CQR probability at 1 in 2.125 x 10^22. or 212.5 times higher than White gives. Jehu The calculation I made is from memory and comes from numbers in another paper by White. The 10^12 number of parasites is true ONLY for those who become seriously ill with malaria---and probably end up dying. That represents just a percentage of all those who are infected. That's why for (b) I used the figure of 4 to 5% of the world-wide number infected. As I also pointed out, though, this number becomes around 10^19 parasites in the whole world in any one calendar year----which is in contradiction of White's number of 10^17 for the total number of parasites. So, there's a discrepancy. And, Behe's number could possibly be criticized for being too high. But let's note two things: first, the number is still extremely high, and, second, the 1 in 10^20 is a number that White comes up with, NOT calculating "probabilities", but just a "statistical fact" regarding CQR. Behe's response to Coyne at his Amazon.com blog fully states his argument. And fully illustrates the dilemna that fair-minded Darwinists face. Fred Hoyle, in "Mathematics of Evolution", using a type of path-integral formula right out of quantum mechanics, calculated that the most that evolution could do in sexual species is to move one or two steps in one direction or other. Snoke and Behe, using a computer model, calculate that the population size, and the time period needed, to bring about a double a.a. switch in coding DNA are both extremely high. Now, per White and his malarial statistics, a CCC is 1 in 10^20, leading to the mathematical conclusion that about the most that RM+NS can do is bring about two a.a. changes. As they say: Strike Three! You're Out!!! PaV

Looks like I made an error, the population of sick people is 4.25 x 10^8. So, according to my calculations, that would put the de novo CQR probability at 1 in 2.125 x 10^22. or 212.5 times higher than White gives. Jehu

PaV: “Then, per White, (a) is 10^12; (b) world-wide 350-500 x 10^6 people have malaria each year; but of those a small percentage get seriously ill. If we assume four to five percent of those with malaria are seriously ill, that gives 2 x 10^7; (c ) is 5 years (ten developments of CQR in 50 years): Thus, (10^12) x (2 x 10^7) x 5 = (approx) 10^20.” According to the World Health Organization, 350 to 500 million people became ill with Malaria in 2003. So 425 million is a good estimate for the number of people who get sick with Malaria each year. That is 4.25 x 10^7 sick people, with a population of 10^12 each, per year for 50 years. With those numbers, I calculate a total population of 2.125 x 10^22. If we allow CQR has occurred de novo 10 times (it has actually only been discovered 4 or 5 times I believe) then the probability of CQR occuring de novo 1 in 2.125 x 10^21. That is 21.25 times higher than the number that White gives. You really have to allow very slim chances of selection to dent those numbers. Since we are only talking about sick people, that criticism doesn't really hold water. Jehu

magnan,

Each case is the result following a long process of immune system attack of any clones of parasites that happened to acquire CQR, and differential fitness selection (at least prior to administration of the drug). So it seems the actual rate of parasites developing CQR must be higher than the 1 in 10^20 figure, if White derived the 10^20 number the way suggested by PaV.

I don't think White used the same numbers as PaV. One thing is clear, with his over 400 publications on the topic, White clearly understands that odds against a de novo occurance being selected and that consideration taken into his calculation. Just read some of White other papers that have been linked to in this thread if you are interested. You also have to consider the population that the figure is calculated from. Sick people. It is not calculated from the population of all malaria in the world. Sick people have compromised immune systems. I pointed this out in post #71. In the 2004 paper, White states, "Taken together, the balance of evidence strongly favors acute symptomatic infection as the source of de novo antimalarial resistance.” Additionally, White calculates the number of times CQR has arisen at "less than 10" not "5". As I previously pointed out in post #102, a massive NIH study found only four historic instances of CQR occuring de novo. This would mean that White allowed for roughly double the number of de novo instances than have actually been found. From what I have read, this is consistent with population genetics probabilities of a trait being selected under strong selective pressure. In stating the estimated number of times CQR has occurred de novo, White cites to, Su, X., Kirkman, L.A., Fujioka, H., and Wellems, T.E. 1997. Complex polymorphisms in an approximately 330 kDa protein are linked to chloroquine-resistant P. falciparum in Southeast Asia and Africa. Cell. 91:593-603. Unfortunately, I don't have access to that article. It would be important to see how the "less than 10" number is justified. So if anbody can post the referenced portion of the article it would be appreciated. Jehu

PaV: "Then, per White, (a) is 10^12; (b) world-wide 350-500 x 10^6 people have malaria each year; but of those a small percentage get seriously ill. If we assume four to five percent of those with malaria are seriously ill, that gives 2 x 10^7; (c ) is 5 years (ten developments of CQR in 50 years): Thus, (10^12) x (2 x 10^7) x 5 = (approx) 10^20." I have been trying to understand this issue despite JAM's distasteful and unclear posts. This may get me banned from this blog, but I just have a need to understand the truth of these issues, whatever it is and regardless of the consequences. I say this as an ID advocate. (c) above is the rate of observed clinical resistance developing somewhere in the world. Each case is the result following a long process of immune system attack of any clones of parasites that happened to acquire CQR, and differential fitness selection (at least prior to administration of the drug). So it seems the actual rate of parasites developing CQR must be higher than the 1 in 10^20 figure, if White derived the 10^20 number the way suggested by PaV. White's statement in his 2004 paper in which he furnished the 10^20 figure: "...This suggests that the per-parasite probability of developing resistance de novo is on the order of 1 in 10^20 parasite multiplications." The critic was claiming that White's statement was poorly written, and that it should have been as follows: "...This suggests that the probability of a parasite both developing resistance and passing on this resistance to produce clinically observed resistance is on the order of 1 in 10^20 multiplications." If this interpretation is correct, then White's figure of 1 in 10^12 for development of "one mutation" resistance (based on atovaquone) was also poorly stated in the paper and would be rewritten in the same way, so the actual per-parasite probability for atovaquone resistance must be much larger. I really want to discover why this reasoning is incorrect. At the same time I have already posted on how this issue needs to be put in perspective. magnan

JAM: You have invoked recombination as a mechanism that increases the probability of getting two mutations simultaneously. However, malarial parasites, though eukaryotic, when reproducing in humans, do so asexually, thus eliminating the possibility of recombination except for a few meiotic cycles in the mosquito. As to "simultanaeity": White published in another paper a number for the world-wide number of malarial parasites for any given year. His number, i.e., the total number of malarial cells in any one given year, is 10^17. If a simultaneous mutation involving two a.a.s is 10^20, then this would be expected once in every thousand years. This strongly suggests that the mutations occur sequentially, and not simultaneously. Now, having said that, I think if we follow what Behe writes on p. 57: "Nicholas White of Mahidol University in Thailand points out that if you mulitply the number of parasites in a person who is very ill with malaria (a) times the number of people who get malaria per year (b) times the number of years since the introduction of chloroquine (c ), then you can estimate that the odds of a parasite developing resistance to chloroquine is roughly one in ahuncred billion billion." Then, per White, (a) is 10^12; (b) world-wide 350-500 x 10^6 people have malaria each year; but of those a small percentage get seriously ill. If we assume four to five percent of those with malaria are seriously ill, that gives 2 x 10^7; (c ) is 5 years (ten developments of CQR in 50 years): Thus, (10^12) x (2 x 10^7) x 5 = (approx) 10^20. But I have to remark that (a) x (b) is greater than 10^17; which gives me pause. Anyway, however nature wanted to work it, it took a hundred billion billion (per White) malarial cells (=organisms) to come up with two mutations that developed resistance to chloroquine. It seems to me Behe's argument still stands. PaV

JAM argued a strawman: Behe was very clear in specifying a single mutation shifting two amino acids that is required to BECOME (not to simply BE clinically) resistant.

Baloney. At some point people run out of patience having to deal with a participant who argues his case by willfully attributing things to ID proponents which they did not say. JAM showed incompetence in interpreting Fidock, but still did not relent. He made a rather ridiculous interpretation of that paper, and when I called him on it he finally relented. That didn't stop him from spewing out more garbage that I just wasn't willing to deal with. Many thanks the admins for dispensing with jam. scordova

JAM: thank you for your comments. I'll try to answer some of your points: 1) Why should Behe discuss the origin of the Vpu protein any more than the origin of any other functional protein in HIV or in any other living system? It's you who have cited the Vpu in the discussion. As far as I know, there is no reason that this particular protein is not designed, like, in my opinion, any complex functional protein in any living system. It's darwinian theory which, without any evidence, declares that all of them have evolved by mechanical, step by step, RM + NS. So, I ask again: are you aware of a detailed, step by step molecular path of evolution for that protein (or for any other) by darwinian means? I think I have answered your question, will you please answer mine? 2) You say: "I’m sorry, but teaching a basic genetics course is beyond the scope of a blog comment". ButI have not asked a course. I only ask that you clarify, as briefly and clearly as possible, your statement that: due to recombination, “the probability of getting two independent mutations together is many orders of magnitude higher”. I just miss the explanation of your quantitative statement. Would you be so kind as to say something more? After all, even if this is only a blog, everyone here is trying to make his points clear to others. Except you, maybe. 3) I appreciate your clarifications about resistance not being binary, and I agree with you on that point. I don't think that changes very much the problem, as you seem to affirm. The fact remains that you cited an article which is only about in vitro manipulations, and has probably scarce connections with what happens in vivo. I would like you to specify better your thoughts. You criticize Behe for having made assumptions you don't agree with. Behe thinks, deriving that from White, that the much lower incidence of CQR compared with other resistances is due to the fact that at least two aminoacid mutations are needed to confer a true "in vivo" selection of the mutant. You say, if I understand well, that that's not true, or at least is not supported enough by facts. Your explanation seems to be that one mutation is enough for CQR, and that its lower incidence (extremely lower) is due to reversal for loss of fitness. Have I understood well? In that case, even if your objections to Behe should have some basis (which I don't agree), I don't see how your hypothesis is more supported by facts. It seems to me at least as speculative as Behe's. So, unless you have stronger evidence that your explanation is right, I think we can discuss both explanations, and decide for ourselves which one fits data better. In any case, I don't understand how your alternative explanation could in any way undermine the general thesis of Behe's book: if the parasite, in all that time, has not even been able to generate a mutation for which at least two substitutions are necessary before getting an advantage, but has only managed single point selectable mutations of the "burning the bridge" type, then Behe's thesis is stronger than ever. I would like to restate that Behe's point is that no new true biochemichal machine, no new protein interaction, has been developed by the parasite to adapt to the strong constraints of S hemoglobin and/or of drugs. Against that thesis, you have cited an article about point mutations, "CTL Escape Mutations", which are exactly of the "burning the bridge" type, and you have not answered my previous note about that, and you have also cited, out of the blue, the Vpu protein, whose pertinence I still miss. Again, with Behe, I affirm that anything too complex to have evolved by step by step darwinian mechanisms is best explained by design. Unless, obviously, you can show us the detailed molecular path through which it should have reasonably evolved. Again: the problem is not if CQR mutation is simpler than Behe assumes. Anyway, Behe believes that CQR has been achieved by darwinian mechanisms. The problem is that CQR, and similar, perhaps simpler, mutations, are the best thing that malaria parasite could generate, under strong darwinian constraints, by darwinian mechanisms. Nothing of the kind of a flagellum, to be clear, or of a complex new enzyme, protein cascade, or molecular machine. That is Behe's thesis, and to that you should give a reasonable answer. gpuccio

JAM is no longer with us. Patrick Caldon is no longer with us. DaveScot

Jehu: Given the large populations of malaria in sick people, that doesn’t help you make much headway against Behe’s arugment, you still need two mutations to get CQR.

Behe's argument has zero evidence to support it. There is substantial evidence against it. It's telling that you misrepresent a hypothesis as a fact. Science isn't about arguments, it's about evidence derived by testing hypotheses. It's especially important to produce new evidence, something that Behe doesn't do. JAM

Jehu: What is claimed by Behe is that two mutations are required to confer CQR.

How do you derive "two mutations" from "single mutation," Jehu?

...there is no significant positive selection until both are present.

There's zero evidence to support that claim. There is substantial evidence against it, which I cited.

White already factored in the various reasons while CQR will not be selected every time it emerges, including the role of the immune system.

White clearly did, but Behe doesn't. He attributes it all to mutation rates. That's the difference. Which one is the expert?

As for your polio argument, some experiments can be performed in vitro and do not need to be in vivo. It all depends on the context of what you are attempting to determine. So I cannot respond to your polio argument.

You're not even close. Of course you cannot respond, because doing so would open your eyes to how you've been fooled by Behe!

I said your cystic fibrosis argument was laughable because you are comparing an allele that is under negative selection with an allele that is under positive selection.

1) The CFTR mutant allele is under BOTH positive selection (in heterozygotes in Second and Third World nations, who are more resistant to diarrhea) and negative selection (in homozygotes worldwide). 2) The CQR alleles are under BOTH positive selection (in treated humans) as well as negative selection (in mosquitoes and untreated humans). 3) In both cases, sexual reproduction acts as a buffer, maintaining polymorphism in the population and "resisting" selection in both directions. 4) The Plasmodium case is much more fluid, because of shorter generation times and more variance in selection pressures. JAM

JAM,

The initial CQR mutations (the first is most likely K76T) have a bigger fitness cost in the absence of drug than the atovaquanone ones do.

Given the large populations of malaria in sick people, that doesn't help you make much headway against Behe's arugment, you still need two mutations to get CQR.

As a scientist, I make predictions. This hypothesis predicts that most, possibly all, of the other 3-7 substitutions in the CQR haplotypes, when isolated in yeast (the method you blow off), will be compensatory and will NOT increase fitness in the presence of CQ, but will instead increase fitness of K76T in the absence of CQ.

Before you get CQR you need a mutation at 76 and 220 or 76, 144, and 160. The other mutations, as has been previously stated, probably confer fitness on the CQR haplotype. Jehu

gpuccio: First of all, I would appreciate if you were so kind to express more clearly and motivate for us ignorant your concept that, due to recombination, “the probability of getting two independent mutations together is many orders of magnitude higher”,

I'm sorry, but teaching a basic genetics course is beyond the scope of a blog comment.

1) You end one of your posts with the following: “Does Behe discuss the evolution of the HIV Vpu protein, or does he conclude that it was added by a designer?

So why not answer my question?

I don’t understand. Are you suggesting that you know a step by step darwinian evolution of the Vpu protein?

Why? Are you suggesting that God designed it in the last century or so? Behe says that his alleged limits require the intervention of a designer. I'm merely asking you, as someone who apparently takes Behe seriously, to tell me which side of that limit Vpu falls on. You can't discuss HIV evolution in any meaningful way without going into detail on Vpu, which does amazing things despite its small size and recent origin, which is true whether it was designed or evolved.

I want to say that I don’t understand your problem with distinguishing between “selective advantage” and “clinically significant CQR”.

It's orders of magnitude, measured both in vitro and in vivo. The concept of differing curves, along with Emax and EC50 is explained very well in the White & Pongtavornpinyo review. You're going to get selective advantages at doses far below the EC50, and plenty of patients will have serum levels below the EC50. Even the patients who follow instructions will at the beginning and end of treatment.

While the two concepts are certainly not the same thing in general, here we are talking of resistance to CQ in an individual infected.

It's a continuum, not a binary characteristic.

The only significant “fitness constraint”, in this context, is resistance to CQ, and it is an advantage only if it is enough to allow infection of the host who is receiving CQ,

It's not binary at all. Go back to my math challenge, an do the math in the presence of CQ dose so small that the relative fitness of the initial CQR mutant is only 1.02. What will be the ratio of mutant to wild-type after 2-3 weeks?

and myultiplication of the parasite in “that” host.

So, in this context, the only true selective advantage is clinically significant resistance to CQ. That's not what White says.

Indeed, if the parasite is resistant enough, it can multiply in the treated host, otherwise it dies.

Its fate depends on two continuous, nonlinear variables: its level of resistance and the concentration of CQ. Sublethal concentrations of CQ are still going to select against wild-type organisms even when they don't kill them. In mosquitoes and untreated patients, the selection is not merely absent, but reversed. For your model to be correct, the fitnesses could never be fractional. You are assuming that the fitness of the mutant is 1 and the wild-type fitness is 0. It just ain't that simple. JAM

Sal: The mistake you made was to assume K76I is sufficient to confer CQR.

No, that wasn't my mistake. The question is whether either K76T or K76I is sufficient to confer ANY selective advantage in the presence of CQ. This is the converse of the math challenge I issued, but which only Patrick had the courage to take up.

But that was not the mutation Behe was speaking of.

Behe was very clear in specifying a single mutation shifting two amino acids that is required to BECOME (not to simply BE clinically) resistant. He explicitly states that "a shift of two amino acids" in a "single mutation" is the stumbling block: “...any single mutation of the kind required for malaria to become resistant to chloroquine–not the easiest mutation, to be sure, but still only a shift of two amino acids–..."

Thank you for the offer of the PDF to the other paper, but even that was a bit of a suspect argument as this was not a study of real malaria but only parts of it inserted in yeast.

Sal, your attempt to trivialize it through strategic omissions is ludicrous. The paper shows that mutating K76 to either I or T AND NOTHING ELSE, results in a substantial increase in CQ transporter activity, the most likely mechanism underlying CQR. This is the only presently available way to test the contribution of each individual change. It is a coisogenic experiment, the best kind.

The issue is whether K76* mutations by themselves are sufficient, and it appears they are not.

No, from these well-controlled experiments, it appears that they are more than sufficient to get the process started by themselves. ------

Behe: The likelihood that Homo sapiens achieved any single mutation of the kind required for malaria to become resistant to chloroquine–not the easiest mutation, to be sure, but still only a shift of two amino acids...

Sal: Was he saying the only abosolute possible route to CQR is via 2 mutations,...

He wrote that a single mutation shifting two amino acids is required.

... or was he referring to the two most important mutations in the primary form of CQR observed and studied so far?

How could he possibly have been referring to TWO mutations if he wrote "SINGLE mutation"?

If you don’t know, say so.

I can say with complete confidence that the phrases "two most important mutations" and "single mutation" cannot possibly be referring to the same things or events.

But let me point out, Behe has made the appropriate qualifiers in other parts of the book, and you should have taken them into account if you were willing to render a charitable reading:

No amount of charity can equivocate between "two most important mutations" and "single mutation," Sal.

Behe: ...different mutations have been found in PfCRT from different regions of the globe…

It does nothing to negate his claim that a single mutation shifting two amino acids is required.

The mutant PfCRTs exhibit a range of changes, affecting as few as four amino acids to as many as eight.

The presence of other changes does nothing to contradict a claim that two changes are required as a single mutation.

What’s this, Behe refers to 4 mutations, and then to 8 mutations for CQ resistance? Apparenlty, the number 2 was referring to the apparent minimal number needed for the CQRs observed,

Nope. Behe specified a "single mutation" and "a shift of two amino acids," and he used the verb "become," reflecting the process, not the end result. Furthermore, clinical CQ resistance is orders of magnitude more than that required to gain a selective advantage in the presence of CQ.

... not that 2 mutations are the only absolute possible route..

He described it as a "single mutation," Sal, that caused "a shift of two amino acids." Why do you put words in Behe's mouth?

JAM, I’m afraid, you’re unwilling to see what Behe is trying to communicate.

Behe is trying to communicate that the rarity of emergence of CQR is primarily the result of a low mutation rate, when in fact, all the experts agree that it is the result of conflicting selection pressures complicated by sexual reproduction and inability to maintain selective pressure in real-world clinical situations.

In light of this, even the 3 amino acid Philippine CQR which I cited does not refute the idea trying to be conveyed. The number 2 was referring to the minimum of 2 mutations needed, not that that was the only path to CQR.

No, Sal, the number 2 did not refer to the minimum of 2 mutations needed. It referred to the number of amino-acid shifts in a single mutation that Behe claims is required. Behe: "single mutation" "required" "a shift of two amino acids"

Well then, how do you account for the fact CQR too longer to evolve than other forms atovaquone.

The initial CQR mutations (the first is most likely K76T) have a bigger fitness cost in the absence of drug than the atovaquanone ones do. As a scientist, I make predictions. This hypothesis predicts that most, possibly all, of the other 3-7 substitutions in the CQR haplotypes, when isolated in yeast (the method you blow off), will be compensatory and will NOT increase fitness in the presence of CQ, but will instead increase fitness of K76T in the absence of CQ. Now, why don't you answer my questions, Sal. 1) If the rarity of emergence of CQR represents a failure of Darwinian evolution, instead of an illustration of MET and population genetics applied to a fitness landscape that not only changes, but reverses, over time, how do you explain the fact that CQR emerges more frequently in regions of low-intensity transmission than in regions of high intensity? 2) Do you see the mechanism of deception in my totally dishonest polio argument? 3) Why don't you do the math to see if a relative fitness of 0.86 in Plasmodium is small or huge? 4) Why don't you do the same problem, assuming only K76T with a fitness of 1.02 in a human who has a low serum concentration (clinically ineffective) of CQ?

You’re persistent, JAM, and dialogue with you has been good practice at rhetoric, but I think you’re not persuading the ID side,

Is that even possible? How can I convince anyone who claims that "two mutations" and "single mutation" refer to the same thing? How can I convince anyone who can't identify the deception in an argument that I specify is deliberately deceptive?

... and I’m not so sure your side is believing what you have to say.

Well, Patrick did the math, and you didn't.

The bottom line is that Darwinism, in light of the very large number trials, was relatively slow to evolve resistance to CQ.

The bottom line is that Behe is trying to misrepresent the delay as being caused by mutation rates, when in fact, it is caused by conflicting selection pressures, which obviously are going to slow Darwinian evolution down. Darwinism doesn't evolve anything; organisms evolve. JAM

JAM, What is claimed by Behe is that two mutations are required to confer CQR. After these two mutations are achieved they are positively selected and subsequent mutations can be sequentially added to improve the fitness of the CQR haplotype. The first two mutations can be sequential or the result of the recombination of two different mutant haplotypes but there is no significant positive selection until both are present. White's estimate of 1 in 10^20 as the per-parasite probability of CQR arising de novo is properly cited by Behe and it is the correct number. White already factored in the various reasons while CQR will not be selected every time it emerges, including the role of the immune system. As for your polio argument, some experiments can be performed in vitro and do not need to be in vivo. It all depends on the context of what you are attempting to determine. So I cannot respond to your polio argument. I said your cystic fibrosis argument was laughable because you are comparing an allele that is under negative selection with an allele that is under positive selection. Jehu

JAM: sincerely, I can't follow the rythm of all your discussions. First of all, I would appreciate if you were so kind to express more clearly and motivate for us ignorant your concept that, due to recombination, "the probability of getting two independent mutations together is many orders of magnitude higher", and the important role of the haplotype concept for that, instead of generivally suggesting that we read genetic textbooks. After all, this is a blog, and for the sake of discussion, you should elucidate explicitly your arguments, so that we can either be convinced and happy, or answer them. For the moment, I will answer a couple of things that you have said explicitly: 1) You end one of your posts with the following: "Does Behe discuss the evolution of the HIV Vpu protein, or does he conclude that it was added by a designer? Would you call this case of observed evolution “not much”? Nature 412, 334-338 (19 July 2001) Evolution and transmission of stable CTL escape mutations in HIV infection Philip J. R. Goulder et al." I don't understand. Are you suggesting that you know a step by step darwinian evolution of the Vpu protein? And what is the meaning of the article you cite? I have checked, and I would definitely say that the escape mutations of HIV which are treated in it, and in several similar articles, are just a perfect example of what Behe extensively discusses, that is point mutations which, by changing an epitope of an existing structure, do provide an advantage towards an aggressor (CTLs). Exactly as in antibiotic resistance. Exactly what Behe very correctly defines "burning bridges", and very clearly states is inthe normal range of what darwinian evolution can do. Have you read TEOE? Besides, some of these point mutations happen with loss of fitness, and reverse after the removal of the constraint (see: HIV Evolution: CTL Escape Mutation and Reversion After Transmission, at the following URL: www.medscape.com/viewarticle/471679_2). Therefore, I don't understand what is the point you are trying to make. 2) I paste here a couple of your comments to my post: My post: "Indeed, I have read the article you cite (Arch Biochem Biophys. 2006 Aug 15;452(2):119-28), and my impression is that it does not demonstrate at all what you say, that the single K76T mutation is sufficient to confer, if isolated, clinically significant CQR." Your comment: "It doesn’t have to confer clinically significant CQR by itself. All it has to do to falsify Behe’s claim is to confer a selective advantage in the presence of CQ." My post: "I would like to add again, anyway, that Behe’s central idea, that if two or more independent mutations have to be present at the same time in an individual to confer a selective advantage,…" Your comment: "But Behe has shown no evidence to support his claim. Why did you lower the goalpost from “clinically significant” above to a mere “selective advantage” here?" I want to say that I don't understand your problem with distinguishing between "selective advantage" and "clinically significant CQR". While the two concepts are certainly not the same thing in general, here we are talking of resistance to CQ in an individual infected. The only significant "fitness constraint", in this context, is resistance to CQ, and it is an advantage only if it is enough to allow infection of the host who is receiving CQ, and myultiplication of the parasite in "that" host. So, in this context, the only true selective advantage is clinically significant resistance to CQ. Indeed, if the parasite is resistant enough, it can multiply in the treated host, otherwise it dies. And believe me, if the parasite multiplies, the host is clinically affected. After all, this is my field... gpuccio

Jehu: You are completely wrong on everything you post.

So when I posted the following, it was wrong? "White himself says that host immunity plays a central role in preventing the emergence and spread of resistance." How was that wrong, precisely? Here's a section heading from the review Behe cites: The central role of immunity in preventing the emergence and spread of resistance in high-transmission settings

Behe doesn’t rely on himself as an authority in Malaria. He relies on N.J. White, who has over 400 publications in the field.

Behe doesn't agree with White. Behe claims that the 10E-20 frequency represents a mutational limit, while White patiently goes through all the relevant factors in detail, such as host immune response and sexual reproduction, both of which Behe omits. Go back to my totally dishonest polio argument above. If you attack my conclusion that animal experiments don't work, I can simply deflect your attack by pretending that you are attacking the authority of the Nobel laurates who did the research instead of attacking my utterly dishonest argument. Would you buy such a deflection as valid? Did you figure out why that argument is dishonest yet?

Your claim that White’s review to article ratio is too high is an unbelievable absurdity that leads me to believe that you are a complete neophyte.

Are you willing to test the sincerity of your belief by putting some real money on it?

White takes into consideration all of the factors you raise and still comes up with a 1 in 10^20 probability of CQR emerging de novo.

And White, unlike Behe, discusses those factors so that anyone who reads his review with an open mind sees the complexity. Behe pretends that it is much more simple than it is.

Your criticism that Behe doesn’t discuss these same factors is therefore pointless.

Behe attributes the rarity of emergence to only one of the factors discussed by White, just as my completely dishonest polio argument attributes all the success of the Nobel laureates' research to their use of human cells in culture. I'm telling you that my polio argument is designed to deceive in precisely the same way, but apparently you can't see the deception. Cognitive dissonance is an amazing thing.

It has been repeatedly explained to you why the evidence best supports CQR requiring two or more amino acid changes from the wild type.

And you clearly aren't bothering to read what I post. It is clear that many changes contribute, and I predict that more will arise in the mutant haplotypes. I am challenging Behe's explicit claim that two are required in a "single mutation." I am challenging any claim that any two have to occur in the same clone, because sexual reproduction recombines them at the same time it preserves wild-type alleles. The selective forces act in different directions, at different times. Behe, through negligent or deceptive omissions, leads his audience to conclude that selection is in one direction only and unchanging over time, so that the reader concludes that this represents a limit to evolution, instead of the complexity of population genetics in sexually-reproducing organisms subject to ever-changing selection pressures. Now, Jehu, can you spot the dishonesty in my polio argument? Above, you said that my argument was laughable, and all you've done is misunderstand and misrepresent my argument. I can tell you up front that my polio argument is dishonest and laughable, and you can't tell me the reason why. JAM

JAM continues to argue: Except that you are ignoring recombination, so the probability of getting two independent mutations together is many orders of magnitude higher.

Well then, how do you account for the fact CQR too longer to evolve than other forms atovaquone. You're persistent, JAM, and dialogue with you has been good practice at rhetoric, but I think you're not persuading the ID side, and I'm not so sure your side is believing what you have to say. The bottom line is that Darwinism, in light of the very large number trials, was relatively slow to evolve resistance to CQ. scordova

JAM, You are completely wrong on everything you post. Your arguments have been thoroughly refuted and yet you repeat them. Behe doesn't rely on himself as an authority in Malaria. He relies on N.J. White, who has over 400 publications in the field. Your claim that White's review to article ratio is too high is an unbelievable absurdity that leads me to believe that you are a complete neophyte. White takes into consideration all of the factors you raise and still comes up with a 1 in 10^20 probability of CQR emerging de novo. Your criticism that Behe doesn't discuss these same factors is therefore pointless. It has been repeatedly explained to you why the evidence best supports CQR requiring two or more amino acid changes from the wild type. You apparently do not understand the arguments because when they are raised you agree with them but then immediately go back to pretending like they were never raised. At this point you are raising one pointless absurdity after another and reverting back to already refuted arguments. Jehu

Bad formatting--the end of the dishonest argument should go like this: viruses; and (c) by identification of the agents grown in tissue culture in virus neutralization tests.” 4) This clearly shows that animal experiments aren’t needed, and that the nonanimal alternative of cell culture gives superior results. JAM

Jehu: It has repeatedly been pointed out to you that when White made his calulation he took into consideration the standard population genetics factors that a de novo mutation for CQR will not be selected when it occurs, including sexual recombination.

I agree completely. For example, White goes into great depth about the role of the host immune response in restricting the spread of resistance, which Behe completely ignores, along with his neglect of sexual reproductive factors.

He takes these factors into consideration and still arives at the number that Behe accurately cites.

"He" being White, yes. But Behe completely misrepresents this probability as the product of only one of those factors (mutation), and claims that two residues must be changed in the absence of evidence supporting his claim, while ignoring evidence that does not support his claim.

As I have pointed out White has over 400 peer review publications on Malaria.

And Behe has zero! Thus, when White discusses the roles of the immune system and sexual reproduction in retarding the spread of CQR and Behe ignores those factors completely, which one has more credibility with you?

You made the utterly trivial argument that 50 of them were reviews. (I am not sure how you know that unless you read them.)

It's easy. They are all identified as reviews, so one simply uses the search term "review." The conflation of reviews with the primary literature (and quoting conclusions with citing the data) is not trivial. In fact, it's one of the best markers for determining whether the quoter's goal is to inform or to deceive.

Regardless 350 articles is still a staggering number and 50 reviews is also impressive.

Yes and no. I'd say his proportion of reviews among his papers is too high. Whether you agree with that or not, when White tells you how the host immune response restricts the spread of CQR in a population, what do you think when Behe ignores that?

With a CV like that, White is undoubtedly one of the World’s leading authorities on the topic. For you to sit here and claim that White doesn’t know the basics of how drug resistance emerges and is selected is absurd.

I haven't claimed anything of the sort. I pointed out that he mislabeled one thing that he correctly labeled elsewhere, probably because he isn't comfortable with the terminology of population genetics.

That is exactly wrong. What you are trying to do is suggest an argument with out actually making one. In this case, the argument you are suggesting is laughable.

Well, since you got the false idea that I was claiming that White doesn't understand how CQR evolves, I'm curious to see what you perceive this argument as being. Let me offer a parallel, totally dishonest argument on a different subject. 1) Animal experiments and tests are misleading. 2) The biggest advance in making a polio vaccine used human cell culture. 3) Here's a quote to prove it: "...the decision was taken to use a mixture of human embryonic skin and muscle tissue in suspended cell cultures in the hope that the virus of varicella might multiply in the cells of its natural host. In this way such cultures were made available while close at hand in the storage cabinet was the Lansing strain of poliomyelitis virus. Thereupon it suddenly occurred to us that everything had been prepared almost without conscious effort on our part for a new attempt to cultivate the agent in extraneural tissue...It is evident that under these conditions significant quantities of virus appear only after a period of eight days and that the maximal yield is obtained between the 12th and 16th days. These findings were soon confirmed and extended: (a) by continued serial passage in human embryonic skin and muscle tissue of this as well as another strain of Lansing virus; (b) by the successful propagation of representatives of Types I and III viruses; and (c) by identification of the agents grown in tissue culture in virus neutralization tests."4) This clearly shows that animal experiments aren't needed, and that the nonanimal alternative of cell culture gives superior results. Jehu, can you spot the defect in the reasoning that makes this a massively, cynically dishonest argument whose sole purpose can only be to deceive? Given your tactics above, is it fair to say that if I pointed out that this argument was totally dishonest, you'd disagree and claim that the argument has to be right because it employs a quote from the ultimate expert in the field--a Nobel laureate? JAM

jerry: I will ask again. Isn’t a main conclusion of the book that given all the reproductive events of the malarial organism, HIV and bacteria that not much has happened to any of these organisms?

I'd call it a contention, not a conclusion. You'd need to review the evidence exhaustively to come to such a conclusion, and Behe clearly hasn't done that. Does Behe discuss the evolution of the HIV Vpu protein, or does he conclude that it was added by a designer? Would you call this case of observed evolution "not much"? Nature 412, 334-338 (19 July 2001) Evolution and transmission of stable CTL escape mutations in HIV infection Philip J. R. Goulder et al. JAM

JAM

The reason that resistance takes so long to emerge is the same reason that we will never be rid of cystic fibrosis alleles–population genetics, particularly the ability of sexual reproduction to maintain polymorphism in a population.

That is exactly wrong. What you are trying to do is suggest an argument with out actually making one. In this case, the argument you are suggesting is laughable. Jehu

JAM; It has repeatedly been pointed out to you that when White made his calulation he took into consideration the standard population genetics factors that a de novo mutation for CQR will not be selected when it occurs, including sexual recombination. He takes these factors into consideration and still arives at the number that Behe accurately cites. As I have pointed out White has over 400 peer review publications on Malaria. You made the utterly trivial argument that 50 of them were reviews. (I am not sure how you know that unless you read them.) Regardless 350 articles is still a staggering number and 50 reviews is also impressive. With a CV like that, White is undoubtedly one of the World's leading authorities on the topic. For you to sit here and claim that White doesn't know the basics of how drug resistance emerges and is selected is absurd. Jehu

gpuccio: I don’t see how sexual reproduction could increse the probabilities, and make Behe’s number “exponentially wrong”.

I suggest a genetics textbook, then. They cover recombination in detail. Sexual reproduction both makes it more likely to recombine independent events and less likely for them to be transmitted, both human->mosquito and mosquito->human. It's much more complicated than Behe's presentation suggests it is.

I can’t answer your objection at this regard, because that is not my field and I could easily be wrong.

It's not Behe's field either, so I I don't understand why you accept his oversimplification of White's position by ignoring the host immune response and sexual reproduction. White himself says that host immunity plays a central role in preventing the emergence and spread of resistance. Whom do you believe?

I think magnan has well answered some of your objections, but I agree that the subject is complex.

Do we agree that Behe oversimplifies it?

Indeed, I have read the article you cite (Arch Biochem Biophys. 2006 Aug 15;452(2):119-28), and my impression is that it does not demonstrate at all what you say, that the single K76T mutation is sufficient to confer, if isolated, clinically significant CQR.

It doesn't have to confer clinically significant CQR by itself. All it has to do to falsify Behe's claim is to confer a selective advantage in the presence of CQ.

I would like to add again, anyway, that Behe’s central idea, that if two or more independent mutations have to be present at the same time in an individual to confer a selective advantage,...

But Behe has shown no evidence to support his claim. Why did you lower the goalpost from "clinically significant" above to a mere "selective advantage" here?

...that should offer a self-evident impossibility for causal role of RM + NS in most significant biological information building.

Why? We've seen that evolution can co-opt an existing pathway in a very short time. The reason that resistance takes so long to emerge is the same reason that we will never be rid of cystic fibrosis alleles--population genetics, particularly the ability of sexual reproduction to maintain polymorphism in a population.

Finally, I can’t share your indignation at citing from reviews, especially in a book aimed at the general public.

The problem is in quoting from reviews instead of citing the actual data. I don't see why the target audience excuses the failure to go to the data, as books aimed at the general public cite the data all the time. Quoting from reviews enables obfuscation.

Anyway, Behe gives many direct references in the notes.

But not on the central point in what Behe calls his "centerpiece example."

Reviews, metanalyses, and in general intellectual discussions about data are, in my opinion, a very important part od scientific knowledge.

Reviews are used to introduce people to a field. They are not used as sources of quotes by practicing scientists looking to bolster their position in a controversy.

Peer reviewed lab work is not necessarily the hallmark of scientific truth.

Then what is the hallmark, if not evidence (by evidence, I don't mean only lab work; it includes field work)? And why do reviews cite the data published in these manuscripts, and why is their primary function to point interested parties to them? --------

Patrick Caldon: the probability of a sequence of mutations occurring in a particular generation cannot be calculated directly from the number of distinct times a particular phenotype (e.g. CQR) been observed to arise and become predominant in a population in the real world. These are chalk and cheese; the one question is essentially molecular biology, the other essentially population genetics.

Precisely. It's clear that White understands this but none of that is conveyed in Behe's book. JAM

mangan: This is an excellent conceptualization, and I think it explains why “simultaneous” or “sequential” for the mutations conferring CQR are irrelevant to the issue.

It's not even close to excellent, because it completely ignores the role of meiotic recombination in combining substitutions in a haplotype.

Under these conditions the net probability is still approximately the product of the two mutation frequencies even when the two occur sequentially.

Except that you are ignoring recombination, so the probability of getting two independent mutations together is many orders of magnitude higher.

This analysis neatly fits with White’s estimates of mutation frequencies for resistance.

How does a discrepancy of almost 1000-fold become a neat fit in your estimation? Your whole position consists of promoting White's mislabeling of something (a single time), because despite his expertise, he clearly is uncomfortable with the language of population genetics.

He considers multiple mutations in his description of Table 1:

No, that's only one of the things that White lists in his Table 1: Factors determining the probability of selection of de novo antimalarial drug resistance 1. The frequency with which the resistance mechanism arises 2. The fitness cost to the parasite associated with the resistance mechanism 3. The number of parasites in the human host that are exposed to the drug 4. The concentrations of drug to which these parasites are exposed (i.e., the doses used and pharmacokinetic properties of the antimalarial drug or drugs) 5. The pharmacodynamic properties of the antimalarial drug or drugs 6. The degree of resistance (the shift in the concentration-effect relationship) that results from the genetic changes 7. The level of host defense (nonspecific and specific immunity) 8. The simultaneous presence of other antimalarial drugs or substances in the blood that will still kill the parasite if it develops resistance to one drug (i.e., the use of combinations) All these factors contribute to the 10E-20 frequency. You and Behe simply pretend that most of these factors do not exist.

Behe realized this and just describes the two as being necessary in one parasite at the same time without specifying whether this acquisition was simultaneous or sequential.

No. Behe clearly stated that a "single mutation" causing a "shift of two amino acids" is "required": "...single mutation of the kind required for malaria to become resistant to chloroquine–not the easiest mutation, to be sure, but still only a shift of two amino acids..."

White notes that mutation frequencies are considerably higher in vitro than in vivo.

Wrong again. White carefully qualifies the in vivo frequency as an APPARENT mutation rate, because of all those factors listed above. Why is it that when White writes something you like, you claim, "It is what White says it is," but when he writes something you don't like, you omit the qualifications that he includes that you don't like?

This would be because in the body as opposed to in vitro the single mutation is not sufficient for expanding clonal selection against the drug.

Here's what White (and nobody else) says: "Mutations may be associated with fitness disadvantages (i.e., in the absence of the drug they are less fit and multiply less well than their drug-sensitive counterparts). Another factor that may explain the discrepancy between in vitro and much lower apparent in vivo rates of spontaneous mutation is host immunity....there is only a 2–3% chance that the genetic event causing resistance would arise in the antigenically variant subpopulation that will expand to reach transmissible densities." So, mangan, is that 50-fold reduction a part of White's "per-parasite probability of developing resistance de novo" that Behe touts as a simple mutation rate?

For this it appears to need several additional mutations, which have to occur either simultaneously or soon thereafter in the clone of the first mutation.

False; neither requirement exists. You are simply denying the existence of sexual reproduction in Plasmodium.

Thus, again, the net frequency of in vivo CQR acquisition is approximately the product of the mutation frequencies (or even lower), and corresponds closely to the estimates used by Behe.

Read the list White offers. Those are the factors that contribute to the low frequency of in vivo CQR acquisition, and Behe is trying to pretend that all those factors represent only one: a mutation rate. Then, he takes that rate and applies it to humans, despite the fact that we can and have measured mutation rates in humans. Why extrapolate from the complex population genetics of primitive eukaryotic parasites? Why not look at the mutational events underlying mammalian evolution to see if Behe's limits hold? Do small genetic changes mediate large morphological changes? JAM

jam I also recall you saying that it has been empircally established that the single point mutation rate in prokaryotes is ~10^9. You then go on to argue that the possibility of two simultaneous mutations is ~10^18 and then ask how it became 10^20 in malaria parasites. The answer is simple and I'm surprised you (or anyone else I read here) didn't know the answer. The answer is that each single point copy error can be to any one of four nucleotides (ACTG). The CQR mutation requires a specific one of those four possibilities while the other three are useless for CQR. Thus it raises the improbability of the right mutation occuring by a factor of 4 or ~4*10^9. Since the 10^9 number is approximate to begin with (it varies by environment, genetic loci, and species) a multiplication by four could very conceivably bump the base rate up to ~10^10. The 10^20 number for CQR was empirical, not calculated. Not surprisingly it roughly agrees with the calculated probability of two specific simultaneous point mutations. It also agrees nicely with the empirical observation of atovaquone resistance (AQR) arising in 10^10 malaria parasites. AQR requires only one single point mutation (again the right one of four possibilities). In your ancillary argument that CQR resistance could be two sequential point mutations - that is correct - it theoretically *could* be. Facts however say it doesn't happen that way. Virtually every person treated with atovaquone ends up with a resistant bug. If two sequential mutations could confer CQR then we should expect that it would arise at about half the rate of AQR. But in fact CQR is exceedingly rare. That CQR requires two simultaneous mutations is thus a virtually indisputable inference from real world data. Behe is on very solid ground here and your critique has now been shown to be fatally flawed. I suggest you move on to another avenue of attack. DaveScot

jam I recall you asking, more than once IIRC, where Behe got a particular number: 10^20. I glanced at his Amazon blog where he addresses it in a response to Jerry Coyne.

Here is where Professor Coyne and other Darwinist reviewers really miss the boat and overlook the considerable power of the malaria results. The number I cite, one parasite in every 1020 for de novo chloroquine resistance, is not a probability calculation. Rather, it is a statistic, a result, a data point. (Furthermore, it is not my number, but that of the eminent malariologist Nicholas White.) I do not assume that “adaptation cannot occur one mutation at a time”; I assume nothing at all. I am simply looking at the results. The malaria parasite was free to do whatever it could in nature; to evolve resistance, or outcompete its fellow parasites, by whatever evolutionary pathway was available in the wild. Neither I nor anyone else were manipulating the results. What we see when we look at chloroquine-resistant malaria is pristine data — it is the best that random mutation plus selection was able to accomplish in the wild in 1020 tries. Let me elaborate that last point. The fact that de novo chloroquine resistance is observed to be an event of frequency 1 in 1020 means that mutational events of greater frequency are of little help, because events of greater frequency would have been expected to occur many times in the same time interval. For example, if a single point mutation such as K76T alone in PfCRT in the wild were sufficient to confer chloroquine resistance, then resistance would occur de novo in virtually every person treated with chloroquine, as it does in almost every person treated with atovaquone. In 1020 parasites that single mutation would have been expected to have occurred about 1010 times or more. What’s more, every other possible single point mutation, at every position of the parasite’s genome, would also be expected to have occurred roughly the same number of times. And enormous numbers of other types of mutations — deletions, insertions, gene duplications, and more — in every gene of the parasite, would also have occurred. The result: a very few mutations helped the parasite a bit; the overwhelming number of mutations did not help at all.

DaveScot

Now, let's revisit something I asked repeatedly.

JAM said: what matters is that Behe claims that two changes are required simultaneously.

And then I asked:

1. one particular kind of CQR malarial strain 2. ANY conceivable CQR malarial strain. 3. You don’t know how to answer Why don’t you answer the question? Afraid to respond. Or will you just quote-mine passages out of Behe’s book to avoid answering? It’s not that hard a question.

And then you said:

I do know what you’re asking. That’s why I don’t simply say #3.

Ok, let's try again with a different approach. When Behe says:

The likelihood that Homo sapiens achieved any single mutation of the kind required for malaria to become resistant to chloroquine–not the easiest mutation, to be sure, but still only a shift of two amino acids

Was he saying the only abosolute possible route to CQR is via 2 mutations, or was he referring to the two most important mutations in the primary form of CQR observed and studied so far? If you don't know, say so. But let me point out, Behe has made the appropriate qualifiers in other parts of the book, and you should have taken them into account if you were willing to render a charitable reading:

different mutations have been found in PfCRT from different regions of the globe… The mutant PfCRTs exhibit a range of changes, affecting as few as four amino acids to as many as eight. .... some other mutations in some other proteins are thought to contribute to chloroquine resistance, none are nearly as effective as that in PfCRT Michael Behe

What's this, Behe refers to 4 mutations, and then to 8 mutations for CQ resistance? Apparenlty, the number 2 was referring to the apparent minimal number needed for the CQRs observed, not that 2 mutations are the only absolute possible route.. JAM, I'm afraid, you're unwilling to see what Behe is trying to communicate. In light of this, even the 3 amino acid Philippine CQR which I cited does not refute the idea trying to be conveyed. The number 2 was referring to the minimum of 2 mutations needed, not that that was the only path to CQR. PS You're participation here shows a great degree of resolve and valor, and I respect that, but it is badly misplaced. You're quibbling over words, and mis-reading what is said. You badly misread Behe's work and you tried to criticize his ideas with your misreading of the Fidock paper. Your valor is commendable, but your making one mistake after another at this point. scordova

JAM wrote: Sal, My mistake. I missed that they had put K76I it into the Dd2 context.

Gracious of you to acknowledge your egregious error. The mistake you made was to assume K76I is sufficient to confer CQR. But that was not the mutation Behe was speaking of. Thank you for the offer of the PDF to the other paper, but even that was a bit of a suspect argument as this was not a study of real malaria but only parts of it inserted in yeast. The issue is whether K76* mutations by themselves are sufficient, and it appears they are not. scordova

I will ask again. Isn't a main conclusion of the book that given all the reproductive events of the malarial organism, HIV and bacteria that not much has happened to any of these organisms? And that their number of reproductive events dwarf the number of reproductive events of all mammals since their inception. JAM did not answer the question but deflected to another argument being the center piece of the book. jerry

Patrick Caldon, One thing that gets a comment held up is a link. The longer the link the more likely the comment will not make it through. There are times when I cannot even reference another thread at UD. jerry

Some of my comments seem to be ending up in the bit-bucket, so it seems I must be violating the comment policy and being boring. I shall try to be less boring! Patrick (with no last name), "what is the source of novelty" is an excellent question. But there are two questions here: * what is the source of "novel adaptive function of the organism occurring as a result of a new allele arising, and subsequently fixing into a sizable sub-population", which will of course depend on what the adaptation is to, what the population is and what environment the population is exposed to. and, * "what is the source of new alleles", which is a question about molecular biology. JAMs consistent point (and argued with much more technical knowledge than I or anyone else here possess) is that the probability of a sequence of mutations occurring in a particular generation cannot be calculated directly from the number of distinct times a particular phenotype (e.g. CQR) been observed to arise and become predominant in a population in the real world. These are chalk and cheese; the one question is essentially molecular biology, the other essentially population genetics. From my limited knowledge of genetics JAMs point seems an eminently sensible, and entirely consistent with standard references/ textbooks on the subject. I'm not qualified to say that he's right, but for instance his careful insistence on the importance of the notion of haplotype has given me a lot of insight into these kinds of mechanisms which I did not previously have. Patrick Caldon

JAM: my note: The two mutations could well appear sequentially, for instance one in a single plasmodium and the second in its immediate descendants. your comment: I agree…or they can be combined in a haplotype by recombination during sexual reproduction. All these things happen billions of times more frequently than one in 10E20. So we agree that Behe’s number is exponentially wrong. No, I don't think we agree on that last concept. The fact is that, if two independent mutations have to be simultaneously present in an individual plasmodium to confer resistance, be selected and expand, then the probability of the event is the product of the two single probabilities, however and whenever and wherever the two single mutations happende. I don't see how sexual reproduction could increse the probabilities, and make Behe's number "exponentially wrong". I appreciate your discussion about the fundamental "if". Indeed, Behe's argument, or at least his main argument in the book, rests on the assumption that both mutations are necessary for CQR. I can't answer your objection at this regard, because that is not my field and I could easily be wrong. I think magnan has well answered some of your objections, but I agree that the subject is complex. Indeed, I have read the article you cite (Arch Biochem Biophys. 2006 Aug 15;452(2):119-28), and my impression is that it does not demonstrate at all what you say, that the single K76T mutation is sufficient to confer, if isolated, clinically significant CQR. But again, this is not my field, and so I will let this discussion to others. I would like to add again, anyway, that Behe's central idea, that if two or more independent mutations have to be present at the same time in an individual to confer a selective advantage, then the probabilities of the event are extremely low, whatever the random mechanism by which the single mutations arise. Given the very complex nature of practically every new biochemical function, that should offer a self-evident impossibility for causal role of RM + NS in most significant biological information building. Finally, I can't share your indignation at citing from reviews, especially in a book aimed at the general public. Anyway, Behe gives many direct references in the notes. Reviews, metanalyses, and in general intellectual discussions about data are, in my opinion, a very important part od scientific knowledge. Peer reviewed lab work is not necessarily the hallmark of scientific truth. gpuccio

gpuccio: "In other words, the chances of having the two mutations sequentually in the same individual plasmodium line are not so different from those of having them arise “simultaneously”, that is in the same cell,in a short time, because anyway the clone of the original mutation remains a small clone, does not expand, and can well die, and so the chances of having the specific second mutation in that specific clone are extremely low, and should be multiplied to the equally low chances of the first mutation to obtain the total chances of the double event in a same individual." This is an excellent conceptualization, and I think it explains why "simultaneous" or "sequential" for the mutations conferring CQR are irrelevant to the issue. Under these conditions the net probability is still approximately the product of the two mutation frequencies even when the two occur sequentially. This analysis neatly fits with White's estimates of mutation frequencies for resistance. He considers multiple mutations in his description of Table 1: "If the resistance mechanism is multigenic then this represents the frequency of the parasite becoming resistant and thus it is the product of the individual mutation frequencies." He doesn't mention simultaneous or sequential because in this situation it is irrelevant. Behe realized this and just describes the two as being necessary in one parasite at the same time without specifying whether this acquisition was simultaneous or sequential. White notes that mutation frequencies are considerably higher in vitro than in vivo. This would be because in the body as opposed to in vitro the single mutation is not sufficient for expanding clonal selection against the drug. For this it appears to need several additional mutations, which have to occur either simultaneously or soon thereafter in the clone of the first mutation. Thus, again, the net frequency of in vivo CQR acquisition is approximately the product of the mutation frequencies (or even lower), and corresponds closely to the estimates used by Behe. magnan

I wonder what survival advantage is incurred by snideness. Do the mutations need to be simultaneous, or is it likely to be conferred sequentially, through single mutations for sneering and sarcasm? Apollos

The relative fitness of CQR strains of malaria, in the absence of Chloroquine, has been calculated at .76 and .86, depending on the study.

Then why not cite those studies directly, instead of a review that cites them? You wouldn't be denigrating them as insignificant if you did.

The population of CQR malaria in a single individual necessary to produce illness is at least 10^7. At that point, the gametocytes carrying the CQR genes have reached transmissible densities and will not get lost due to sexual reproduction in the mosquito.

Why are you assuming that the infection and population growth occurred in the presence of CQ?

White, who has published over 400 peer review articles on malaria, appears to have already taken all of these factors into consideration in making his calculations.

False on both counts: 1) Fifty-two of his papers are reviews. They are rarely peer-reviewed. In fact, Behe is avoiding the primary, peer-reviewed literature in favor of quoting from a review. Can you guess why? 2) White clearly tells you that not all of the factors can be quantitated or estimated.

It is also being claimed that Behe is assuming two simultaneous mutations are required to confer CQR. This is false. What Behe is saying is that a minimum of two mutations are required before CQR arises, not that the mutations must occur simultaneously.

Behe is very clear: “The likelihood that Homo sapiens achieved any single mutation of the kind required for malaria to become resistant to chloroquine–not the easiest mutation, to be sure, but still only a shift of two amino acids–the likelihood that such a mutation could arise just once in the entire course of the human lineage in the past ten million years, is minuscule–of the same order as, say, the likelihood of you personally winning the Powerball lottery by buying a single ticket.” We know that the probability of getting a substitution at a specific nucleotide is 10E-9. The probability of getting two substitutions simultaneously at two specific nucleotides is the square of that, or 10E-18. Where does Behe derive 10E-20 if, as he claims, he is looking at mutation frequencies? Are you saying that Behe was in error when he wrote this? --------

gpuccio: It seems to me that the question should be viewed as follows: what Behe means is, as Jehu correctly states, that for resistance to CQR to arise, it is necessary that at least two independent mutations be present in the same individual plasmodium. That does not mean that the two mutations must arise simultaneously, in the sense that they must happen “at the same time”, or even in the same individual plasmodium.

That's not what Behe wrote, though.

The two mutations could well appear sequentially, for instance one in a single plasmodium and the second in its immediate descendants.

I agree...or they can be combined in a haplotype by recombination during sexual reproduction. All these things happen billions of times more frequently than one in 10E20. So we agree that Behe's number is exponentially wrong.

But the point is that neither one nor the other confer resistance to CQR if isolated,..

But the hallmark one, K76T, has been isolated experimentally, and you're wrong: Arch Biochem Biophys. 2006 Aug 15;452(2):119-28 Functional reconstitution of purified chloroquine resistance membrane transporter expressed in yeast. Tan W, Gou DM, Tai E, Zhao YZ, Chow LM. This is an isogenic experiment, as clean as it gets.

... and therefore even if they appear sequentially, the clone with the first mutation cannot in any way be selected, fixed or expanded, although it certainly has the same chances to survive as any other CQR sensitive clone (not many, if CQR is administered).

But that assumption is wrong. You're also trying to pretend that CQR/CQS is simple and binary. Real people have all sorts of different CQ concentrations, changing over time.

...That is not only a theoretical calculation, but is strongly supported by the observation of the extreme rarity of CQR resistance (two mutations)

Not two mutations. Two haplotypes with multiple mutations. Why won't anyone think about the definition of "haplotype" here? --------

jerry: But I read the book through once and it seemed to me the main conclusion was that the malaria eukaryote represented more reproductive events than the reproductive events in a lot of classes such as mammals since the beginning of their existence. And given this enormous number of reproductive events, very little has happened to change the organism compared to the zillions of changes that have taken place in mammals.

But you can only derive that probability if you assume that two simulaneous changes are required in a single event, and that event is still ~1000x more frequent than Behe's "centerpiece example" of 10E-20.

So the argument over when and how a couple mutations may or may not have happened is a sideshow argument over irrelevant minutiae. Or am I not understanding something which is always possible. If I am misunderstanding the connection of the minutiae with the overall conclusion then maybe someone could discuss the connectionns.

Behe calls it the "centerpiece example" of the book, so if you call its derivation "irrelevant" and a "sideshow," you are clearly disagreeing with Behe. --------

magnan: JAM: “Since the detection of it (spontaneous resistance mutation)...

This is the second time you've inserted "mutation" where it wasn't used. Why not argue the actual evidence? The antecedent of "it" is not a "spontaneous resistance mutation," it is a resistance HAPLOTYPE that contains multiple substitutions. Why do the experts use the word "haplotype," mangan?

Since the inference is based on White’s research results that he clearly described,

Behe's inference was definitely not based on White's research results. It was cynically based on a quote Behe took out of context from a REVIEW. That tells you just how weak it is.

...then both of White’s papers and probably many of his others are also fatally flawed, despite his apparently being accepted as an expert in the field with a large number of published research papers.

You're citing a review, not a research paper.

You need to point this out in the scientific arena by a letter to the Royal Society to be published in its Proceedings.

Why? I don't object to anything in that review.

White et. al. first present an analysis of the genetic events conferring drug resistance including the estimates of de novo per parasite resistance mutation frequencies with the 10^-19 figure for CQR (Table 1).

Wrong, mangan. You omitted the qualifications White included in the title: "Approximate per-parasite frequencies for genetic events (mutations or gene amplifications) which lead to the emergence of clinically significant drug resistance of Plasmodium falciparum in vivo." So you omitted "lead to," "clinically significant," and the plural on "mutations." How come? Wishful thinking? Why not take a peek at Figure 3 while looking at Table 1?

The next paragraph is on “De Novo selection of resistance” in which he discusses the probability that the genetic event conferring resistance will be selected for and expand in the body population. After this he covers all the factors reducing individual parasite survival probability and whether resistant strains reach transmissable populations. This includes reverse selection in the mosquito part of the life cycle.

Yes. This is just one of the reasons why the probability of emergence in a population is so much lower than the mutation frequency.

White and Pongtavornpinyo are analyzing de novo emergence of antimalarial drug resistance, as they say predominantly in areas of low malaria transmission.

No, it's a review, not an analysis. Do you understand the purpose of reviews?

This means spontaneously occurring resistance in individuals, not transmitted resistance.

They introduce you to the reasons why resistance in individuals is rarely transmitted. You apparently missed that. And if they are only discussing resistance in individuals, what is the relevance of the rate of transmission? Face it, mangan, they are doing epidemiology and population genetics, not analyzing mutation frequencies, which are much more easily measured for humans as rates of sporadic dominant lethals.

Thus the Table 1 estimate of “per parasite resistance mutation frequency” for CQR is clearly distinguished from host selection and transmission factors. It is what White says it is.

Yet at least one of the authors says that it is something different in the title. Do you realize how pathetic this is, mangan? How do you know that Pongtavornpinyo didn't write the column heading and White didn't write the far more accurate title for the table? Do you see that if you are reduced to quoting a column heading as evidence, you are deliberately ignoring the actual evidence?

Later in the paper White continues to make clear his (to you misguided) conclusions: “By contrast, for drugs such as chlorquine or artemisinin, the genetic events conferring resistance are much rarer (they may have happened only a few times in the case of chlorquine, and significant resistance has not yet been detected for artemisinin). Assuming an equal distribution of probabilities throughout the life cycle, the genetic event is likely to take place in only a single parasite at the peak of infection.”

I don't claim that he's misguided. I also don't claim to be certain that White wrote that. And it's clear that they are writing about clinically-significant resistance, which involves multiple, clinically-undetectable steps.

Notice that he...

He? Funny, I noticed that there were two authors, not one. I just sent the first draft of a ms out to my coauthors, and they are finding plenty of ambiguities. Of course, because this paper, unlike a review, will be peer-reviewed, it is much more important to correct them.

What do you think the following quote means from the White paper (in section 5. Transmission intensity and the selection of resistance)?: “In low-transmission areas the majority of malaria is symptomatic and selection therefore takes place in the context of treatment.”

It means that the emergence of resistance is so rare because it doesn't emerge in areas of high transmission. It doesn't help you to address my question, which is about whether any real-world malaria outbreak has had all patients treated throughout infection with CQ. Why not just admit that the answer is no? Show some courage and do the math. 10 CQR sporozoites (fitness 0.86) and 10 CQS sporozoites (fitness 1) infect an untreated human. Three weeks later, what is the typical number of merozoites in the blood and what is the calculated ratio of CQR to CQS merozoites? JAM

The Other Patrick,

Most modern biologists apparently don’t subscribe to the view “that evolution is the result of mutation and natural selection” as you state, but would state rather that adaptation (as opposed to evolution) is the result of natural selection.

The real question is what is the source of novelty. Two years ago at the World Summit on Evolution it appeared that the "old camp" of Neo-Darwinists was still strong but that other camps were growing. I asked around at the time if anyone knew of a study or poll on what modern biologists believed but no one knew. I'd still like to know the size of the various "camps". Debating with someone who is in the same camp as Dawkins is very different from, say, Margulis or MacNeill. They're all coming from a different point of view on "how evolution really works". This leads to confusion when someone in the two sides debating doesn't know the other's position. It'd be like if someone here were to attack DaveScot with an argument commonly targeted at Creationists. Oh, and there has been much discussion about neutral theory in the past. Try googling "neutral site:www.uncommondescent.com". Here's one from Sal: https://uncommondesc.wpengine.com/intelligent-design/what-are-the-speed-limits-of-naturalistic-evolution/ Patrick

DaveScot, It's not hard to see that the vast majority of mutations in eukaryotes will be neutral, since they will occur in regions which don't code for anything or regulate anything. In this way we can confidently predict that the vast majority of mutations will be neutral. Synonymous mutations will all be neutral. Indeed I understand that the majority of coding mutations are neutral or nearly neutral, and the empirical studies which have been done indicate a preponderance of neutral mutations, a good many detrimental mutations but a small but still noticeable number of beneficial mutations. For instance this paper found that 12% of a sample of random mutations on E. Coli were beneficial: http://www.pnas.org/cgi/content/abstract/98/20/11388 The big problem is in fact: how do the neutral mutations become fixed? I'm no expert in this area as I've only read a couple of textbooks on this material, but Kimura and Ohta described how neutral and nearly neutral (i.e. in some cases mildly detrimental) mutations can and necessarily will become fixed over time. Indeed it's possible to predict the degree of heterozygosity which comes from drift (i.e. related to the number of unfixed mutations hanging around) and predict it will be a function of the rate of mutation and the population size. Most modern biologists apparently don't subscribe to the view "that evolution is the result of mutation and natural selection" as you state, but would state rather that adaptation (as opposed to evolution) is the result of natural selection. In summary, we can confidently predict that the majority of random mutations will be neutral or nearly neutral, and have good theoretical models which have been empirically tested to show how both these mutations become fixed in populations. I've suggested this to you before, that you'd do very well to read an undergraduate textbook on evolution, and get a better grasp of the theory you're criticizing. Try Mark Ridley's book, "Evolution", it explains how the neutral and nearly neutral theories work very well, and then discusses in depth how it matches up with empirical data. Patrick Caldon

JAM: "Since the detection of it (spontaneous resistance mutation) is only possible after many life cycles of selection for alternating with selection against this, your inference is fatally flawed." Since the inference is based on White's research results that he clearly described, then both of White's papers and probably many of his others are also fatally flawed, despite his apparently being accepted as an expert in the field with a large number of published research papers. You need to point this out in the scientific arena by a letter to the Royal Society to be published in its Proceedings. White et. al. first present an analysis of the genetic events conferring drug resistance including the estimates of de novo per parasite resistance mutation frequencies with the 10^-19 figure for CQR (Table 1). The next paragraph is on "De Novo selection of resistance" in which he discusses the probability that the genetic event conferring resistance will be selected for and expand in the body population. After this he covers all the factors reducing individual parasite survival probability and whether resistant strains reach transmissable populations. This includes reverse selection in the mosquito part of the life cycle. White and Pongtavornpinyo are analyzing de novo emergence of antimalarial drug resistance, as they say predominantly in areas of low malaria transmission. This means spontaneously occurring resistance in individuals, not transmitted resistance. Thus the Table 1 estimate of "per parasite resistance mutation frequency" for CQR is clearly distinguished from host selection and transmission factors. It is what White says it is. Later in the paper White continues to make clear his (to you misguided) conclusions: "By contrast, for drugs such as chlorquine or artemisinin, the genetic events conferring resistance are much rarer (they may have happened only a few times in the case of chlorquine, and significant resistance has not yet been detected for artemisinin). Assuming an equal distribution of probabilities throughout the life cycle, the genetic event is likely to take place in only a single parasite at the peak of infection." Notice that he is referring to the "genetic event" as occurring in a single parasite probably at the peak of infection (because of the higher population numbers). Notice that this event "May have happened only a few times" (for CQR since introduction of the drug). This is the occurrence of the mutations conferring CQR. Magnan: So during an outbreak RM (and other sources of random variation) + NS in infected humans can be expected to vastly predominate over any reverse selection during the mosquito reproductive life cycle. JAM: Why? Can you point me to a malaria outbreak in the real world in which all the infected humans were treated with CQ? What do you think the following quote means from the White paper (in section 5. Transmission intensity and the selection of resistance)?: "In low-transmission areas the majority of malaria is symptomatic and selection therefore takes place in the context of treatment." magnan

I have a question. I have not followed all the detailed wrangling over the timing and sequences of mutations. This is not my background and I hadn't the time to concentrate on who is consistent or not and what the book actually says. But I read the book through once and it seemed to me the main conclusion was that the malaria eukaryote represented more reproductive events than the reproductive events in a lot of classes such as mammals since the beginning of their existence. And given this enormous number of reproductive events, very little has happened to change the organism compared to the zillions of changes that have taken place in mammals. So the argument over when and how a couple mutations may or may not have happened is a sideshow argument over irrelevant minutiae. Or am I not understanding something which is always possible. If I am misunderstanding the connection of the minutiae with the overall conclusion then maybe someone could discuss the connectionns. jerry

I have received TEOE juast a few days ago, and I have already almost finished it. I am really very, very enthousiastic about the book. It is, in my opinion, a very important book, clear, precise and deep, full of pertinent information, balanced, and very well written. I really shoul catch up with all the discussions here. For the moment, I woul like to add a brief comment about the problem of the "two mutations", simultaneous or not, which seems to be at the center of many posts, including the last one from Jehu. It seems to me that the question should be viewed as follows: what Behe means is, as Jehu correctly states, that for resistance to CQR to arise, it is necessary that at least two independent mutations be present in the same individual plasmodium. That does not mean that the two mutations must arise simultaneously, in the sense that they must happen "at the same time", or even in the same individual plasmodium. The two mutations could well appear sequentially, for instance one in a single plasmodium and the second in its immediate descendants. But the point is that neither one nor the other confer resistance to CQR if isolated, and therefore even if they appear sequentially, the clone with the first mutation cannot in any way be selected, fixed or expanded, although it certainly has the same chances to survive as any other CQR sensitive clone (not many, if CQR is administered). In other words, the chances of having the two mutations sequentually in the same individual plasmodium line are not so different from those of having them arise "simultaneously", that is in the same cell,in a short time, because anyway the clone of the original mutation remains a small clone, does not expand, and can well die, and so the chances of having the specific second mutation in that specific clone are extremely low, and should be multiplied to the equally low chances of the first mutation to obtain the total chances of the double event in a same individual. That is not only a theoretical calculation, but is strongly supported by the observation of the extreme rarity of CQR resistance (two mutations) compared to the much higher frequency of other resistances to other antimalaric drugs (one mutation). gpuccio

The relative fitness of CQR strains of malaria, in the absence of Chloroquine, has been calculated at .76 and .86, depending on the study. The population of CQR malaria in a single individual necessary to produce illness is at least 10^7. At that point, the gametocytes carrying the CQR genes have reached transmissible densities and will not get lost due to sexual reproduction in the mosquito. White, who has published over 400 peer review articles on malaria, appears to have already taken all of these factors into consideration in making his calculations. It is also being claimed that Behe is assuming two simultaneous mutations are required to confer CQR. This is false. What Behe is saying is that a minimum of two mutations are required before CQR arises, not that the mutations must occur simultaneously. Jehu

Patrick Caldon I think it's misleading to say mutations are random with relation to phenotype. Random is synonymous with unpredictable. Saying random mutations are unpredictable is like saying state lottery results are unpredictable. The state wouldn't have lotteries if the results were unpredictable. They don't know precisely who will win and lose but they can confidently predict losers will vastly outnumber winners. If there's one chance in 10 million of winning a given lottery you can point to any individual ticket and predict it won't win with a confidence level of 99.99999 percent! Random mutations are on a par with that. We can confidently predict that any given random mutation won't be beneficial. If you scrutinize Darwin's Origin of Species you'll find two places in it (Chapters 1 and 5 IIRC) indicating Darwin believed in Lamarckian inheritance of acquired characters. If Lamarck was right then it becomes much more credible that evolution is the result of mutation and natural selection. The crux is that Lamarckian mutations aren't random but rather are direct responses to environmental pressure. If mutations are indeed random, and we know that random mutations are rarely beneficial, and even a beneficial mutation in organisms with long reproductive cycles and few offspring stand little chance of becoming fixed due to slightly higher selection value, then it's reasonable to seriously question whether random mutation is the true source of novelty that natural selection acts upon. DaveScot

Phinehas: Is that similar to how the lack of randomness in coin flips and die rolls is largely predicted by physics?

No, it's very different, as I explained and you predictably ignored.

So then, what would count as a random process once we’ve set the standard for randomness so high? Or was that the point?

Well, you set the standard at PURELY random, remember? Again, mutations aren't random wrt location, direction, type, etc. They are random wrt fitness, so that can't meet any rational definition of "purely random," especially when you're applying that label to a process that includes selection, which isn't random by any criterion.

Jehu pointed out earlier: “the relative fitness of CQR strains is .76 or .85, so it is not immediately deselected for lack of fitness in an environment without selective pressure for CQR.”

First, Jehu didn't cite any data for that, and the speed depends on how you define "immediately." Why don't you try a calculation, then? Fitness is per generation. If a mosquito infects you with 10 CQR sporozoites and 10 CQS sporozoites. The proportion of CQR after n generations is then calculated as: (0.5)(0.86)^n Remember, the n is an exponent. How many generations are there in a single human host? And which one of us is engaging in wishful thinking in claiming that 0.86 isn't a significant reduction in fitness?

So, what is the relative fitness of atovaquone resistant strains?

It's the ratio of fitnesses with/without drug, not the absolute value. You really need to think about the math before you do a victory dance. The relative fitnesses will be different at each stage of the life cycle, with the wild card of sexual reproduction thrown in.

magnan made a similar point: “Concerning the rapidly reversing fitness landscape argument. White says the fitness factor of the CQR strain is 75%-85% that of the wild type. This is weak selection compared to the very strong selection for CQ resistance in a multiplying population in a sick human.

1) White doesn't say that anywhere in the JCI review. 2) What White says doesn't matter. What matters is what the data show. Besides, if you want to hold up White as the ultimate authority, he doesn't share Behe's hypothesis about simultaneous mutations. 3) The only fitness paper I found was about pfmdr, the secondary locus, not the primary one (pfcrt), with a relative fitness of 0.75 in an isogenic in vitro experiment with four substitutions: Hayward, Rhys, Saliba, Kevin J. & Kirk, Kiaran pfmdr1 mutations associated with chloroquine resistance incur a fitness cost in Plasmodium falciparum. Molecular Microbiology 55 (4), 1285-1295. 4) Guess how many days of culturing a 50:50 mix of Plasmodium carrying the resistant and sensitive haplotypes it took to reduce the level of resistant ones to an undetectable level. 5) Even if the haplotypes with 4-8 substitutions had a relatively high fitness, I'm talking about the initial steps. Obviously, if resistant bugs have swept the population, many of those substitutions (how does White think they got put together in haplotypes, btw?) help with fitness in the absence of the drug.

In addition, the population in a human victim is vastly greater than in the mosquitoes.

The number is not the critical thing--it's the number of generations, which is an exponent for the fitness. What's (0.86)^1000, for example? Is the number of generations in EACH human host typically more than 1000? Are all humans treated with drugs? What's the fitness of these haplotypes through sexual reproduction? For that matter, does Behe even bother to tell you about the sexual reproduction?

So during an outbreak RM (and other sources of random variation) + NS in infected humans can be expected to vastly predominate over any reverse selection during the mosquito reproductive life cycle.

Why? Can you point me to a malaria outbreak in the real world in which all the infected humans were treated with CQ? And in every life cycle, there's a bottleneck in mosquitoes, with differential fitness at multiple steps, including competition during sexual reproduction, all occurring in the absence of CQ.

Given these points, your assertion seems awfully speculative to me, but I’d be happy to consider any math or data you have to back up the claim that the difference between every third person and only 10 persons in the last fifty years can be explained solely by appealing to differences in relative fitness.

Solely? That's a straw man. You seem to have forgotten the multiple genes involved and the complexity of the haplotypes. How were the haplotypes assembled? What does White say? Do you realize that ethical scientists, when they disagree with dogma, cite the papers with data, and don't quote passages from reviews to try to fool people into thinking that the author of the review agrees with the ethical maverick?

That sounds like wishful thinking to me.

Sounds like projection to me. Why won't you be brave and answer my questions? JAM

Phineas, As a broad analogy, (and I'm a non-expert in this area) think about what happens when you throw several dice. If you throw one, the probabilities of 1,2,3,4,5,6 are equal. If you throw three, there is a good chance of the sum being 9,10,11, or 12 but a very low chance of the sum being 3,4,17, or 18. I believe the point of the word "randomness" in this context is "not-correlated with the fitness of the phenotype". The mutations which happen just happen, without having an eye to what the ultimate effect of the mutations are on the critter. The mutation probabilities are on the other hand correlated with other things; e.g. the organism's exposure to mutagenic processes and substances, the particular place on the chromosome, and so on. Patrick Caldon

JAM: Their lack of randomness is largely predicted by genetics, chemistry and biochemistry. Is that similar to how the lack of randomness in coin flips and die rolls is largely predicted by physics? So then, what would count as a random process once we've set the standard for randomness so high? Or was that the point? Phinehas: Can you please explain the difference in these spontaneous resistance rates? JAM: Easily. Substitutions that confer some degree of CQR decrease fitness in the absence of CQ much more than the substitutions that confer atovaquone decrease fitness in the absence of atovaquone. Jehu pointed out earlier: "the relative fitness of CQR strains is .76 or .85, so it is not immediately deselected for lack of fitness in an environment without selective pressure for CQR." So, what is the relative fitness of atovaquone resistant strains? magnan made a similar point: "Concerning the rapidly reversing fitness landscape argument. White says the fitness factor of the CQR strain is 75%-85% that of the wild type. This is weak selection compared to the very strong selection for CQ resistance in a multiplying population in a sick human. In addition, the population in a human victim is vastly greater than in the mosquitoes. So during an outbreak RM (and other sources of random variation) + NS in infected humans can be expected to vastly predominate over any reverse selection during the mosquito reproductive life cycle. So an alternating fitness landscape doesn’t explain the rarity of CQR." Given these points, your assertion seems awfully speculative to me, but I'd be happy to consider any math or data you have to back up the claim that the difference between every third person and only 10 persons in the last fifty years can be explained solely by appealing to differences in relative fitness. That sounds like wishful thinking to me. Phinehas

CJYman, Their lack of randomness is largely predicted by genetics, chemistry and biochemistry. For example, tandem repeats expand and contract at incredibly high frequencies (10E-5) by recombination. The sequences between tandem repeats are lost by recombination. Triplet CAG repeats (for example, those that cause Huntington's disease) expand by polymerase "stuttering." Transitions occur more frequently than transitions. Chromosomes break more frequently at some places than others, giving translocations. Does that answer your question? JAM

Hello JAM, Just for the sake of educating myself, if mutations are not purely random, which natural law or pattern do they follow? CJYman

Phinehas: Can you please explain the difference in these spontaneous resistance rates?

Easily. Substitutions that confer some degree of CQR decrease fitness in the absence of CQ much more than the substitutions that confer atovaquone decrease fitness in the absence of atovaquone. The most likely scenario is that for CQR, some of the substitutions merely "correct" some of the structural distortions caused by the CQR-conferring substitutions (both K->T and K->I are huge changes). This makes very clear experimental predictions: when the substitutions in resistant haplotypes are mixed and matched in all possible combinations in experimental systems, some will increase CQR while some will increase fitness of the resistance haplotype in the absence of CQR. Now, since I answered your question, how about answering mine: 1) If terminology is more important than data, why is the term "haplotype" used instead of "allele"? 2) Why are there papers on linkage disequilibrium? Does Behe offer a clue? 3) Why choose Plasmodium to extrapolate to human mutation rates instead of using actual human mutation rates?

Are not both arising from the same reversing fitness landscapes?

Yes, but they differ in degree--more accurately, the ratio of fitness relative to wild-type with the drug to without the drug.

Wow. Way to keep your eye on the science and the data. :rolleyes:

Keeping our eyes on the science, Behe's hypothesis (which he misrepresents as fact) suggests radically different strategies for both drug choice and design than does sequential acquisition (particularly including recombination). If he really believes that two substitutions must occur simultaneously, he should be working in his lab instead of doing book tours.

…it seems to me that assuming two SIMULTANEOUS mutations are needed instead of just two SEQUENTIAL mutations puts the accomplishments of random mutation in a better light given the White quote.

It's not about putting things in a better light in the context of a quote from someone who doesn't use accurate genetics terminology. It's about figuring out what happened and how we might be able to save millions of lives, most of them belonging to children.

In other words, if only two SEQUETIAL mutations are needed to confer CQR, and yet it has only arisen spontaneously less that ten times in the last fifty years, doesn’t that say even less about what purely random forces are capable of accomplishing?

There's nothing in evolutionary theory that's "purely random," remember? Mutations are only random with respect to fitness, not "purely random." Selection isn't random at all. If this were "purely random," how is it that we see may of the same substitutions arising independently? Isn't that the antithesis of randomness, much less pure randomness?

Look, I can see that you really want the dispute to be about sequential vs. simultaneous,...

The post was about Miller. That was the precise nature of his criticism of Behe, wasn't it?

...but for me, the ten successes in fifty years is the real point, and those numbers don’t change no matter how much smoke is blown to cloud the issue.

Behe is blowing the smoke here.

I find it extremely refreshing to see evolutionary biology discussing real-world data like this representing what evolutionary processes are actually able to accomplish over a large number of generations in a purely natural setting and not one subject to intelligently designed lab experiments and wishful thinking.

In my real scientific world, the ecological and experimental data complement each other. Neither replaces the other. We won't know which substitutions in CQR haplotypes do what until someone does the experiments in the lab. Would you like to bet some money on what will be found?

I can see how it would be extremely frustrating for Darwinists to have to deal with this sort of real-world data without being able to bring their refined speculative powers to bear.

Maybe you can offer a useful definition of "Darwinist" in this context. Are you suggesting that the authors of these malaria papers are ID proponents, or agnostic about evolutionary theory? Are you suggesting that I am an evolutionary biologist? That Miller is? That we follow a man instead of the evidence? Think about it.

Evolution happens.

So does denial.

And when we look at the evolution that happens in the real world, it turns out that it is very different to the evolution that happens in the imaginations of Darwinists.

In what way is it different? Since you invoked the "purely random" straw man twice, I'd like to see some evidence that you've made an effort to look at the scientific output of scientists like Miller and me.

To me, this is the real take-away from the discussion.

God forbid you might abandon the "purely random" canard. That would represent real intellectual progress. JAM

Phinehas: Perhaps I am just not understanding the argument here, but it seems to me that assuming two mutations are needed instead of just one… JAM: You are completely misunderstanding the dispute here. The question is not whether two mutations are needed, the dispute is whether they must occur SIMULTANEOUSLY and not SEQUENTIALLY. OK. Let me rephrase. ...it seems to me that assuming two SIMULTANEOUS mutations are needed instead of just two SEQUENTIAL mutations puts the accomplishments of random mutation in a better light given the White quote. In other words, if only two SEQUETIAL mutations are needed to confer CQR, and yet it has only arisen spontaneously less that ten times in the last fifty years, doesn’t that say even less about what purely random forces are capable of accomplishing? Look, I can see that you really want the dispute to be about sequential vs. simultaneous, but for me, the ten successes in fifty years is the real point, and those numbers don't change no matter how much smoke is blown to cloud the issue. I find it extremely refreshing to see evolutionary biology discussing real-world data like this representing what evolutionary processes are actually able to accomplish over a large number of generations in a purely natural setting and not one subject to intelligently designed lab experiments and wishful thinking. I can see how it would be extremely frustrating for Darwinists to have to deal with this sort of real-world data without being able to bring their refined speculative powers to bear. Evolution happens. And when we look at the evolution that happens in the real world, it turns out that it is very different to the evolution that happens in the imaginations of Darwinists. To me, this is the real take-away from the discussion. Phinehas

JAM, I note that you keep passing by this: magnan: Spontaneous resistance to atovaquone occurs to about every third sick person. Spontaneous resistance to CQ has only appeared about ten times total in the last 50 years. Can you please explain the difference in these spontaneous resistance rates? Are not both arising from the same reversing fitness landscapes? JAM: The papers you and Jehu are quote-mining were written by people who are trying to save millions of human lives, something that Behe will never attempt. Wow. Way to keep your eye on the science and the data. :rolleyes: Phinehas

Jehu: This in vitro study of CQR found that,”The association of the K76T mutation with chloroquine resistance was not clear. The mutation was detected in resistant and susceptible samples, suggesting that additional factors are involved in chloroquine resistance."

There's no doubt that additional ones are involved. The issue is Behe's assumption that two must occur simultaneously. There's no evidence supporting it.

The researchers further stated,...

What they stated is not the issue. What they showed is the issue. In the real world, many people don't take their medicine often enough or long enough. If you're trying to determine whether a change at K76 has to occur simultaneously with a second change, you have to look at low CQ concentrations. Of course, these researchers weren't trying to address Behe's unsupported assumption that two substitutions had to occur as a single event. They didn't do a dose-response curve.

"...the K76T mutation was also detected in 15 of 17 samples classified as susceptible to chloroquine in vitro.”

What was the classification criterion used?

This is consistent with a mutation at 76 and at least one other position being required for CQR.

That depends on how you define it, which is why quote-mining is so misleading. The issues here are: 1) whether Behe's assumption that two substitutions must occur simultaneously is supported by the evidence, and 2) Whether looking at genotypes after literally billions of generations of on/off selection represents the mutation rate. Whether two changes are REQUIRED for clinically-defined CQR is irrelevant to #1, as single changes would need to be tested for resistance to low concentrations of CQ. If both substitutions HAD to occur to get to resistance, neither can confer ANY resistance to CQ at even the lowest doses.

In fact, White specifies...

In fact, what White specifies or labels doesn't matter. What matters is that to measure mutation rates, you need an absolute, black/white selection. If you want to go with authority (always fallacious in science), BEHE HIMSELF states that the CQR mutants are less fit than wild-type in the absence of CQ, so selection is reversed in the mosquito and in humans not taking CQ. Reversing selection is the antithesis of the simple, absolute selection (like sporadic mutations to dominant lethal alleles in humans) required to infer mutation rates. Think about it: why would Behe choose to extrapolate from Plasmodium to human mutation rates instead of going with measured mutation rates in real humans?

...White very clearly estimated the “per-parasite resistance mutation frequency” of CQR at a very similar 10^19.

Then he was simply wrong in using that term.

In calculating his estimate, White cites to a 1997 paper for the figure of CQR arising 10 times in the last 50 years.

Yes, but it is clear that selection for CQR is intermittent and very complex when it is present. You can't look through the funhouse lens of that kind of selection and claim to be looking directly at mutation rates.

In contrast, a significant 2002 study by the NIH identified only four de novo points of origin for CQR. So there is reason actual rate is even lower. However, as Behe pointed out, it is probably “roughly” 1 in 10^20.

That's not the mutation rate.

Phinehas: Perhaps I am just not understanding the argument here, but it seems to me that assuming two mutations are needed instead of just one...

You are completely misunderstanding the dispute here. The question is not whether two mutations are needed, the dispute is whether they must occur SIMULTANEOUSLY and not SEQUENTIALLY.

...puts the accomplishments of random mutation in a better light given the quote above.

Unfortunately, the job of a scientist is to see if her hypothesis holds up in the worst light by actively testing its predictions, not to cherry-pick observations to support it. The papers you and Jehu are quote-mining were written by people who are trying to save millions of human lives, something that Behe will never attempt.

In other words, if only one mutation is needed to confer CQR, and yet it has only arisen spontaneously less that ten times in the last fifty years, doesn’t that say even less about what purely random forces are capable of accomplishing?

It clearly says that the completely NONrandom forces of selection account for the rarity. Also, please remember that nothing is purely random in evolutionary theory. The mutations are only random in a single way: with respect to fitness.

Given the data, I would think that NDE would come off much better if a rather large number of mutations were required simultaneously for CQR.

Why? The frequency with which it arises is far too HIGH to be accounted for by more than two simultaneous mutations. Did you not notice that up to eight substitutions are found in CQR haplotypes? Say, since specifications seem to be so important to you folks, would you mind explaining why the experts call these "haplotypes" instead of "alleles"? You might try searching for papers that include "linkage disequilibrium" in the title/abstract.

I’m a little stumped about how arguing for fewer mutations is supposed to help NDE’s case.

That's understandable. Miller and I are arguing for sequential mutations, not fewer. Behe is assuming that two must occur simultaneously, with no evidence to support his assumption.

Jehu: If CQR can be achieved by a single step...

Jehu, it's not about whether clinically-defined CQR can be achieved by a single step. It's about whether any of the 4-8 substitutions found in CQR haplotypes (why do they use that word?) had to occur simultaneously.

... and yet has only been selected less than 10 times in 50 years of intense selective pressure in an astronomical number of organisms,

What you're not seeing is that the selection isn't very intense. In fact, Behe himself acknowledges that it gets reversed in the absence of CQ, which happens very often.

then what does that say about the power of natural selection?

What does the power of recombination during sexual reproduction suggest to you?

Mangan: This confirms Behe’s conclusion from the clinically determined resistance in victims and the genetic mutation evidence that CQR requires more than one (at least two) mutations.

The issue, as Miller noted, is whether those mutations occurred simultaneously or sequentially. If they occurred sequentially, given the existence of sexual reproduction in Plasmodium, would the second mutation have had to occur in the initial mutant allele?

JAM: No, because you missed the unsupported assumption that two were required simultaneously. I have shown that the conclusion that at least two were required is well established from research studies. Whether the mutations for CQ resistance were simultaneous or happened sequentially, the clinical rates of appearance of resistance, and White’s estimates of de novo rates clearly indicate the approaching of a limit to NDE as Behe contends.

I suggest that you read the book. Behe is claiming that mutation fails to supply sufficient variation for NS to act upon. The problem with the centerpiece of the book is that his probability has nothing to do with mutation frequencies, and everything to do with reversing selection layered over sequential mutations and recombination during the sexual phase of reproduction (always in the absence of CQ).

...Notice the phrase “lead to the emergence”.

I notice the data more than the phrases. But since you place so much stock in semantics, please explain the use of the term "haplotype" in these papers.

This obviously refers to the probability of the mutation(s) spontaneously occurring in a parasite prior to subsequent selection and expansion of the parasite population in the victim’s body.

Since the detection of it is only possible after many life cycles of selection for alternating with selection against this, your inference is fatally flawed. What does "linkage disequilibrium" obviously refer to? JAM

JAM: "The problem with his conclusion (1 in 10^20) is that it is based on an assumption that two (mutations for CQR) are required, for which he has no evidence." No. This was based first on the clinical data (occurrence of spontaneous resistance in victims) and then supported by the gene mutation data. Spontaneous resistance to atovaquone occurs to about every third sick person. Spontaneous resistance to CQ has only appeared about ten times total in the last 50 years. The 2005 Bray, Martin et. al. (including Fidock) study Behe refers to (page 50) clearly states that K76T is common between the two types of CQR, with the predominant type of CQR being also always accompanied by the position 220 mutation. The other (Philippine) strain has K76T plus two different additional mutations. From the paper: "Note that the critical Lys76 Thr mutation is accompanied by an Arg220 Ser mutation in all CQR lines except for some (the P2a/P2b types) from the Philippines, where this mutation appears to be replaced by the paired Ala144 Thr and Leu160 Tyr mutations..." It would be predicted that the Philippine type of CQR would have a much lower frequency of occurrence, based on there being three rather than two mutations required. The evidence bears that out. The clinical data is that atovaquone resistance (one mutation) is acquired by about one in three victims, and by about one in a billion for CQ (all known occurrences with at least two mutations). If these differences in clinical rates aren't due to numbers of mutations, then what is the factor? If CQR really only requires one mutation, why have no occurrences of this been found? Viana, Machado, Calvosa and Povoa confirm this in their paper "Mutations in the pfmdr1, cg2, and pfcrt genes in Plasmodium Falciparum samples from endemic malaria areas in Rondonia and Para State, Brazilian Amazon Region", at http://www.scielo.br/pdf/csp/v22n12/18.pdf . They state: "There is evidence that chloroquine resistance is a multigenic phenomenon and that the K76T mutation in the pfcrt gene is necessary, but not sufficient, to confer resistance." This confirms Behe's conclusion from the clinically determined resistance in victims and the genetic mutation evidence that CQR requires more than one (at least two) mutations. Magnan: This is because Behe’s estimate is based on the difference between the two numbers (acquisition of atovaquone and CQ resistance), and both numbers have the same bias. The implications of this reasoning for NDE are untouched. JAM: No, because you missed the unsupported assumption that two were required simultaneously. I have shown that the conclusion that at least two were required is well established from research studies. Whether the mutations for CQ resistance were simultaneous or happened sequentially, the clinical rates of appearance of resistance, and White's estimates of de novo rates clearly indicate the approaching of a limit to NDE as Behe contends. I already pointed out that White in the paper cited by Behe clearly states that the estimate of one in 10^20 multiplications is for a new occurrence at random in one parasite in one multiplication, period. Not per parasite just for populations after extensive selection. Thanks Jehu for finding White's later paper "The de novo selection of drug-resistant malaria parasites" (2002) (at http://www.pubmedcentral.nih.gov/picrender.fcgi?artid=1691263&blobtype=pdf ) in which he even more clearly derived a per parasite de novo probability just a little greater than in the 1999 paper (1 in 10^19 multiplications). This figure is described as "per parasite resistance mutation frequency" and is in Table 1, which he describes as "Approximate per-parasite frequencies for genetic events (mutations or gene amplifications) which lead to the emergence of clinically significant drug resistance of Plasmodium falciparum in vivo." Notice the phrase "lead to the emergence". This obviously refers to the probability of the mutation(s) spontaneously occurring in a parasite prior to subsequent selection and expansion of the parasite population in the victim's body. The 1 in 10^12 estimate for atovaquone was also given by White in the same table in the 2002 paper, and has the same definition. magnan

Phinehas, You are exactly right. It appears to be an article of faith amongst Darwinists that everything can be achieved by single step wise mutations. If CQR can be achieved by a single step and yet has only been selected less than 10 times in 50 years of intense selective pressure in an astronomical number of organisms, then what does that say about the power of natural selection? Jehu

The title of the Table that Jehu pulls the 10^19 number indicates that White is most definitely not referring to a simultaneous double mutational event. "Table 1. Approximate per-parasite frequencies for genetic events (mutations or gene amplifications) which lead to the emergence of clinically significant drug resistance of Plasmodium falciparum in vivo." "Mutations" - could be two, could be ten (and the 2003 paper mentions the involvement of multiple genes and epistasis in drug resistance, things not consistent with Behe's overly simplistic portrayal of the situation). It's becoming quite clear that, if Behe is relying on White as an authority, then 10^20 is not even close to being an accurate estimate for the frequency of a simultaneous double mutation. Which brings me back to the question - given this, how does one come close to estimating the "Edge of Evolution" based on chloroquine resistance? Art2

From the White paper: Resistance to chloroquine in P. falciparum has arisen spontaneously less than ten times in the past fifty years (14). Perhaps I am just not understanding the argument here, but it seems to me that assuming two mutations are needed instead of just one puts the accomplishments of random mutation in a better light given the quote above. In other words, if only one mutation is needed to confer CQR, and yet it has only arisen spontaneously less that ten times in the last fifty years, doesn't that say even less about what purely random forces are capable of accomplishing? Given the data, I would think that NDE would come off much better if a rather large number of mutations were required simultaneously for CQR. I'm a little stumped about how arguing for fewer mutations is supposed to help NDE's case. Phinehas

It has been suggested that K76T alone can confer CQR. This in vitro study of CQR found that,"The association of the K76T mutation with chloroquine resistance was not clear. The mutation was detected in resistant and susceptible samples, suggesting that additional factors are involved in chloroquine resistance. The researchers further stated, "we found a correlation between the in vitro response to chloroquine and the mutated sequence at codon 76, which was present in all resistant isolates examined. However, this correlation was far from perfect, as the K76T mutation was also detected in 15 of 17 samples classified as susceptible to chloroquine in vitro." This is consistent with a mutation at 76 and at least one other position being required for CQR. It has also been claimed that Whites figure of 1 in 10^20 is the probability of selection of CQR rather than the probability of a CQR mutation occuring. In fact, White specifies that the figure is a per-parasite figure, stating,"the per-parasite probability of developing resistance de novo is on the order of 1 in 10^20 parasite multiplications." In a paper by White two years earlier titled The de novo selection of drug resistant malaria parasites White very clearly estimated the "per-parasite resistance mutation frequency" of CQR at a very similar 10^19. (Apparently White upped his estimate two years later.) In calculating his estimate, White cites to a 1997 paper for the figure of CQR arising 10 times in the last 50 years. In contrast, a significant 2002 study by the NIH identified only four de novo points of origin for CQR. So there is reason actual rate is even lower. However, as Behe pointed out, it is probably "roughly" 1 in 10^20. Jehu

Sal, My mistake. I missed that they had put K76I it into the Dd2 context. I must have been confusing it with this one, in which either K76T or K76I confer resistance alone: Arch Biochem Biophys. 2006 Aug 15;452(2):119-28 Functional reconstitution of purified chloroquine resistance membrane transporter expressed in yeast. Tan W, Gou DM, Tai E, Zhao YZ, Chow LM. Would you like a PDF? Also, I'm confused by your embrace of a negative result in cultured Plasmodium in culture at #96, while expressing complete disdain for results from culture in #100. The only difference seems to be whether you like the result or not. Is that accurate? BTW, "in vivo" does not mean "in the real world," and "in vitro" does not mean "in laboratory." An experiment done on a mouse in a laboratory is an in vivo experiment. JAM

Why are you playing a semantic game in pretending that when I wrote “only one change to K76,” that I was necessarily referring to K76T?

No need to play semantic games because the refutation of K76T as being a sole source CQ resistance, as you suggested, is shown to be suspect if not completely wrong. Furthermore, in the 2005 paper I cited is co-authored by Fidock. The K76I mutation is not listed in Table 1 of Fidock's earlier paper Mutations in the P. falciparum Digestive Vacuole Transmembrane Protein PfCRT and Evidence for Their Role in Chloroquine Resistance Finally, the 2005 paper assert K76T is the critical mutation in-vivo (in the real world), not K76I which appears apparently only in-vitro (in laboratory ) so far. Further, in the 2000 Fidock paper, the K76I line presumed 7 other mutation were in place (the 106/1 Sudan clone). scordova

Sal, I do know what you're asking. That's why I don't simply say #3. JAM

Sal, Semantic game is right! Why are you playing a semantic game in pretending that when I wrote "only one change to K76," that I was necessarily referring to K76T? JAM

JAM says: So since Behe offers the number of 10E-20, it is closer to #1 than #2.

JAM, You did try to answer the question with that respone, but for the sake of the readers, so that they totally know what you mean. Answer the question simply, not for my sake, but for theirs. You asserted:

what matters is that Behe claims that two changes are required simultaneously.

Does that mean two mutaions are required for: 1. one particular kind of CQR malarial strain 2. ANY conceivable CQR malarial strain. 3. You don’t know how to answer I don't know why it's hard to say #1, #2, or #3? If you don't know what I'm asking, simply say #3. Ok? scordova

JAM confidently asserted at TelicThoughts: Only one change to K76 is required. This has been shown experimentally.

I see some semantic games are being played. In theory K76 might in and of itself do the trick, but because of other factors, it is unlikely. Behe cited the 2000 study by Fidock. Did he cite the study, only then to close his eyes to the data, as JAM insinuates? Consider this 2005 study on the question of the sufficiency of K76T alone: A critical role for PfCRT K76T in Plasmodium falciparum verapamil-reversible chloroquine resistance

While our data indicate a key role for K76T in CQR, it is notable that no fewer than four pfcrt mutations have ever been found in a CQ-resistant isolate (Chen et al, 2003). These other mutations may have been selected to compensate for loss/alteration of endogenous function associated with acquisition of K76T, or may themselves directly contribute to resistance to CQ or other antimalarial drugs. To test whether K76T might itself be sufficient to confer VP-reversible CQR in vitro (which presumably is more favorable than in vivo semi-immune conditions), we employed allelic exchange to introduce solely this mutation into wild-type pfcrt (in GC03). From multiple episomally transfected lines, one showed evidence of K76T substitution in the recombinant, full-length pfcrt locus (data not shown). However, these mutant parasites failed to expand in the bulk culture and could not be cloned, despite numerous attempts. These results suggest reduced parasite viability resulting from K76T in the absence of other pfcrt mutations. This situation is not reciprocal however, in that parasites harboring all the other mutations except for K76T (illustrated by our back-mutants) show no signs of reduced viability in culture.

scordova

Sal: This was referring to a deduction of the approximate likelihood, not an a priori absolute assumption that this is the absolutely the likelihood.

Likelihood of what, exactly? Behe claims that two aa changes are required (even though the data say that only one is) in a single mutation to get resistance. If we apply it to your #1, the probability is 10E-18. If we apply it to your #2, the probability is N x 10E-18, where N is the number of different CQR haplotypes requiring two aa changes in a single mutational event ("malarial strain" is meaningless in this context). So since Behe offers the number of 10E-20, it is closer to #1 than #2. I've responded several times to your question, Sal, and you've falsely accused me of avoiding it. I've also politely asked you to use accurate, unambiguous terms, and you've refused to do so. The bottom line is that Behe claims that two changes are required in a single mutation for resistance, and the experimental data say that only one is required. Behe also cites only ecological data obtained after extensive filtering by selection to draw conclusions about what happens before selection, which is simply indefensible. Please note that there's a huge difference between citing a paper and citing the data in the paper. JAM

JAM quotemines Behe: The likelihood that Homo sapiens achieved any single mutation of the kind required for malaria to become resistant to chloroquine–not the easiest mutation, to be sure, but still only a shift of two amino acids

This was referring to a deduction of the approximate likelihood, not an a priori absolute assumption that this is the absolutely the likelihood. JAM fails to cite the passages which I have cited that would put the appropriate qualifiers on Behe's statement, giving the impression Behe outrightly presumed this was the case from the start. Answer the question JAM. You said:

what matters is that Behe claims that two changes are required simultaneously.

As in: 1. one particular kind of CQR malarial strain 2. ANY conceivable CQR malarial strain. 3. You don’t know how to answer Why don't you answer the question? Afraid to respond. Or will you just quote-mine passages out of Behe's book to avoid answering? It's not that hard a question. scordova

Therefore, if Behe's position is that two specific substitutions are required simultaneously, his number 10E-20 points more to your #1. As I said before, he is ambiguous. If you're going to play NIGYYSOB, I'd appreciate it if you'd use standard genetic terms like "haplotype" and "substitution" and "allele" instead of ambiguous ones like "kind" and "form." JAM

Sal, here are Behe's own words: “The likelihood that Homo sapiens achieved any single mutation of the kind required for malaria to become resistant to chloroquine–not the easiest mutation, to be sure, but still only a shift of two amino acids–the likelihood that such a mutation could arise just once in the entire course of the human lineage in the past ten million years, is minuscule–of the same order as, say, the likelihood of you personally winning the Powerball lottery by buying a single ticket.” We know that the probability of getting a substitution at a specific nucleotide is 10E-9. The probability of getting two substitutions simultaneously at two specific nucleotides is the square of that, or 10E-18. Where does Behe derive 10E-20 if, as he claims, he is looking at mutation frequencies? JAM

JAM misrepresents again: what matters is that Behe claims that two changes are required simultaneously.

As in: 1. one particular kind of CQR malarial strain 2. ANY conceivable CQR malarial strain. 3. You don't know how to answer You're obfuscating a simple question, so I'll pose it to you again. Is it #1 or #2 or #3. If you say, #3, then I'll ask "why do you keep asserting #2 and speaking out of both side of your mouth." scordova

"I know that Behe claims that two are required simultaneously. I know that only one is required. I also know that I never evaded your question." JAM. Where is your proof that only one mutation is needed? Are you making this stuff up?? DarwininianIgnorance

Sal: In that case, you have proven by your own words you have no grounds for insisting that Behe argues a malarial strain will have CQ resistance if and only if it has the 76 and 220 mutations (and by way of extenstion, ONLY 2 mutation combinations).

It doesn't matter which two they are, what matters is that Behe claims that two changes are required simultaneously.

So if you don’t know, it would be advisable to say you don’t know, rather than asserting what you don’t know as a refutation of Behe’s claims.

Behe claims that two are required simultaneously. He doesn't tell his audience that only one is required experimentally. He draws conclusions about mutation rates by looking at what is observed in populations after thousands of generations of both positive and negative selection.

In that case, until you believe you know, kindly refrain from making pronouncemnts that you believe something to be true, when in fact you don’t know.

I know that Behe claims that two are required simultaneously. I know that only one is required. I also know that I never evaded your question.

There are possibly several forms of CQR (Chloroquine Resistant)malarial strains.

If by "forms" you mean "haplotypes," there definitely are several.

One of the most prominent is PfCRT [P. falciparum chloroquine resistance trait].

Indeed, a single substitution in PfCRT is sufficient to confer resistance.

But even with respect to PfCRT resitance alone, not to mention other forms of CQR, Behe writes: different mutations have been found in PfCRT from different regions of the globe… The mutant PfCRTs exhibit a range of changes, affecting as few as four amino acids to as many as eight.

Which is misleading, because a single change of K76 is sufficient.

Later Behe uses the phrase, “almost always”, he doesn’t insist, NEVER.

So where does Behe cite the evidence that demonstrates that two are required?

You just admitted you don’t know this is Behe’s position.

Behe's position is that two changes are required simultaneously. The evidence does not support this assumption. The difference between one and two represents a billion-fold difference in probability, so it matters a lot. JAM

"Do you realize that your assumptions would make the well-studied spontaneous genetic events that occur during human carcinogenesis impossible? You might want to tell friends in Stockholm about this before they give Bert Vogelstein a Nobel Prize." - JAM Hi JAM, I just wanted to point out one or two falsities in your comment above. First of all, there are no well-studied spontaneous genetic events that have beneficent results for the organism. The only time a mutation works to the advantage of a particular organism is if the designer is directing that adaption. Second, your attacks on Behe are sloppy. DarwininianIgnorance

JAM wrote: Then you disagree with Behe, too.

You just admitted you don't know this is Behe's position. You're saying one thing in one post, and then suggesting otherwise in another post. If you don't know for sure what Behe said, then you're in no position to adjudicate whether someone agrees with Behe or not. scordova

In response to my question JAM said: It’s ambiguous, Sal.

Thank you for responding. In that case, you have proven by your own words you have no grounds for insisting that Behe argues a malarial strain will have CQ resistance if and only if it has the 76 and 220 mutations (and by way of extenstion, ONLY 2 mutation combinations). So if you don't know, it would be advisable to say you don't know, rather than asserting what you don't know as a refutation of Behe's claims. In that case, until you believe you know, kindly refrain from making pronouncemnts that you believe something to be true, when in fact you don't know. There are possibly several forms of CQR (Chloroquine Resistant)malarial strains. One of the most prominent is PfCRT [P. falciparum chloroquine resistance trait]. But even with respect to PfCRT resitance alone, not to mention other forms of CQR, Behe writes:

different mutations have been found in PfCRT from different regions of the globe... The mutant PfCRTs exhibit a range of changes, affecting as few as four amino acids to as many as eight.

Later Behe uses the phrase, "almost always", he doesn't insist, NEVER. scordova

mangan: The objection has been made that the clinically derived probabilities of 10^20 and 10^12 from scientific studies are not spontaneous mutation rates, but really reflect probabilities of CQR after periodically varying selection due to body environmental attack and transition outside the body to the mosquito life cycle.

Correct.

This objection does not affect Behe’s conclusion that a slightly more elaborate combination of two primary required specific mutations has only one hundred millionth the chance of spontaneously occurring in a single plasmodium than a single mutation.

The problem with his conclusion is that it is based on an assumption that two are required, for which he has no evidence.

This is because Behe’s estimate is based on the difference between the two numbers, and both numbers have the same bias. The implications of this reasoning for NDE are untouched.

No, because you missed the unsupported assumption that two were required simultaneously.

...Say P. falciparum actually has a spontaneous mutation rate for CQR one billion (10^9) times the 10^20 figure in White’s paper. This would be 1 in 10^(20-9) = 1 in 10^11 multiplications.

Except that we've measured the spontaneous mutation rate directly in many systems (including human cells), and it is consistently 10E-9, 100x higher than your revised estimate. Observed rates from ecological studies are 1-2.5 x 10E-8: Genetics, Vol. 156, 297-304, 2000 Estimate of the Mutation Rate per Nucleotide in Humans Michael W. Nachmana and Susan L. Crowell

...The random mutations or other genetic changes required to build Homo Sapiens from a primitive prehominid would be vastly, astronomically greater than a few single point mutations, so this makes human evolution by a NDE process ridiculous.

On what basis do you assume that the number would be vastly, astronomically greater? For example, are our nervous system connections specified or do they occur by promiscuous connection pruned by selection?

For the same reasons, quibbling over whether two or three mutations are required for CQR, or whether they need to be simultaneous or sequential, is also irrelevant to Behe’s argument.

Then you disagree with Behe, too.

On top of this, Behe didn’t even bother to point out that his extrapolated estimate is based on the P. falciparum genome size of about one hundred million, much less than that of humans. The human genome is about 3 billion nucleotides, making the probability of specific mutational changes by random mutation much less.

No, larger genomes have more mutations, not less. The base frequency is per nucleotide, per replication. Do you realize that your assumptions would make the well-studied spontaneous genetic events that occur during human carcinogenesis impossible? You might want to tell friends in Stockholm about this before they give Bert Vogelstein a Nobel Prize. JAM

scordova: So let me try to rephrase the question. Did Behe argue that a malarial strain will have CQ resistance if and only if it has the 76 and 220 mutations that confer resistance? If you answer #3, then explain why you don’t know how to answer.

It's ambiguous, Sal. He clearly derives his 10E20 from requiring two simultaneous mutations. Whether the second one is in codon 220 or somewhere else doesn't matter. He offers no evidence even suggesting that two simultaneous mutations are required, and doesn't cite the evidence showing that a change in K76 is sufficient to confer resistance in culture: Fidock et al. Mutations in the P. falciparum Digestive Vacuole Transmembrane Protein PfCRT and Evidence for Their Role in Chloroquine Resistance Molecular Cell, Vol. 6, 861–871

jerry: JAM is desperately trying to find some minutae wrong with Behe’s book.

If these are minutae, why does Behe himself write about "the book’s centerpiece example of chloroquine resistance"? JAM

I don’t see how we extrapolate from CQR to limits. May I suggest you read the book and see if that helps your understanding? That's the purpose of the book, after all. DaveScot

A quick reply before firing up the barby ... I don't think anyone is claiming that CQR is a huge evolutionary leap. Also, if the 10^20 figure isn't really correct, then I'm having a hard time seeing how CQR actually relates to, let alone helps define, the "Edge of Evolution". It's just one more example of a multi-genic trait that falls on this side of the "Edge of Evolution". I don't see how we extrapolate from CQR to limits. Art2

Art2, Thank you for that consideration. Let me add however, the bottom line is that CQR does not appear to be a particularly sophisticated evolutionary change. It would be hard to argue that this shows the power of Darwinian evolution to create large scale complexity. It seems to show there are limits. Salvador scordova

Hi magnan, In tossing the number 10^20 out, White seems to be referring to the occurrence of, not a simultaneous double mutation in PfCRT, but rather the coincidence, in a surviving parasite, of resistance-related alleles in three different genes. One allele seems to require two mutations, but it is not clear how many mutations "make up" the other resistance alleles. So, in terms of estimating the "Edge of Evolution", this number (10^20) would seem to be pretty arbitrary, and certainly not related in any strict quantitative way to the simultaneous occurrence of two point mutations. This may be why we can point to examples of at least three "CCC"s occurring, simultaneously and by random processes, at least 30 orders of magnitude more frequently than predicted by Behe's estimates. I think we need to be very tentative about this number. Art2

Art2: "Magnan, I think the three sentences immediately preceding the first quote you give from White’s review are pretty important. And I think they change matters greatly." Please explain. magnan

The objection has been made that the clinically derived probabilities of 10^20 and 10^12 from scientific studies are not spontaneous mutation rates, but really reflect probabilities of CQR after periodically varying selection due to body environmental attack and transition outside the body to the mosquito life cycle. This objection does not affect Behe's conclusion that a slightly more elaborate combination of two primary required specific mutations has only one hundred millionth the chance of spontaneously occurring in a single plasmodium than a single mutation. This is because Behe's estimate is based on the difference between the two numbers, and both numbers have the same bias. The implications of this reasoning for NDE are untouched. This objection, even if correct relative to the meaning of the probability numbers, is also irrelevant to Behe's primary thesis. He shows that there is a vanishingly small likelihood of even a relatively simple combination of specific mutations (as for CQ resistance) occurring in the human population over even millions of years. Say P. falciparum actually has a spontaneous mutation rate for CQR one billion (10^9) times the 10^20 figure in White's paper. This would be 1 in 10^(20-9) = 1 in 10^11 multiplications. Behe does his extrapolation of the CQR estimates to human evolution on page 60 of TEOE. He concludes by simple calculation that "The ratio of humanoid creatures in the last 10 million years to the number of parasites needed for chloroquine resistance is one to a hundred million." This is 1 over 10^8. Even if the number of parasites needed is actually 10^11 not 10^20, then his estimate becomes (one in 10^8) times 10^9 = 10^-1. Still only 1 in 10 for just a relatively simple change of a few substitutions, in 10 million years. Pretty poor performance for NDE. The random mutations or other genetic changes required to build Homo Sapiens from a primitive prehominid would be vastly, astronomically greater than a few single point mutations, so this makes human evolution by a NDE process ridiculous. For the same reasons, quibbling over whether two or three mutations are required for CQR, or whether they need to be simultaneous or sequential, is also irrelevant to Behe's argument. On top of this, Behe didn't even bother to point out that his extrapolated estimate is based on the P. falciparum genome size of about one hundred million, much less than that of humans. The human genome is about 3 billion nucleotides, making the probability of specific mutational changes by random mutation much less. magnan

Sal I don't think I know how to answer it. dougcampo

I see JAM has avoided answering my question, dougcampo as well. scordova

Magnan, I think the three sentences immediately preceding the first quote you give from White's review are pretty important. And I think they change matters greatly. Art2

While I'm on my Berlinski kick. You guys and gals who haven't read this article might enjoy it. http://www.discovery.org/scripts/viewDB/index.php?command=view&id=2450 dougcampo

Here's the quote again from White's paper: "Resistance to chloroquine in P. falciparum has arisen spontaneously less than ten times in the past fifty years (14). This suggests that the per-parasite probability of developing resistance de novo is on the order of 1 in 10^20 parasite multiplications. " Wikipedia definition of "de novo" for evolutionary biology: "De novo mutation: a genetic mutation that neither parent possessed nor transmitted. " So White in the paper cited by Behe clearly states that the estimate of one in 10^20 multiplications is for a new occurrence at random in one parasite in one multiplication, period. Not per parasite for just populations where a parasite with spontaneous CQR won out in a selection process within a human body. Not per parasite for populations resulting from transmission of CQR Plasmodia to a human by a mosquito following continually reversed selection between human infection phases and mosquito life cycles. However, he did enumerate several selection factors that work to slow CQR becoming established in a population in a human body, and becoming transmissible. So in coming up with the 10^20 estimate using in part clinical data White must have accounted for these factors. If he didn't, he would have been guilty of the same mistake Behe is being accused of. Concerning the rapidly reversing fitness landscape argument. White says the fitness factor of the CQR strain is 75%-85% that of the wild type. This is weak selection compared to the very strong selection for CQ resistance in a multiplying population in a sick human. In addition, the population in a human victim is vastly greater than in the mosquitoes. So during an outbreak RM (and other sources of random variation) + NS in infected humans can be expected to vastly predominate over any reverse selection during the mosquito reproductive life cycle. So an alternating fitness landscape doesn't explain the rarity of CQR. Anyway, this "constantly changing fitness landscape" idea is predominantly used by Darwinists to explain why organisms aren't really trapped on local maxima or peaks. Here it is being used to explain an organism being trapped for a long time on a peak. It looks like the dynamically changing fitness landscape argument is going to be an all-purpose "explain away" tool to defeat any and all attacks on NDE. Of course studiously ignoring that it loses all credibility, as if it had any in the first place. A time varying fitness landscape still doesn't get around the probablistic difficulty of credibly assuming that a random walk can move from one isolated "peak" to another. This is just another rhetorical ploy. This debate needs to be kept in perspective. This criticism certainly doesn't challenge any of the many other factors cited by ID advocates. For instance, the No Free Lunch mathematical arguments against NDE including time varying "fitness landscape" explanations, or the general lack of intermediates predicted by NDE, the Cambrian Explosion, Haldane's dilemma, genetic entropy, irreducibly complex biological systems, etc. etc. It doesn't affect Behe's point that mutation (including all sources of random variation including recombination) has to work with pre-existing cellular machinery, so there is a very limited number of things it can do, to paraphrase TEOE page 77. This is another aspect or way of looking at these limitations and is overwhelmingly confirmed by evidence, including the Plasmodium example. At its best this criticises Behe's particular malaria Plasmodium example of the limits of Darwinian evolution. He chose this because of its huge population numbers. It doesn't affect his other examples of such limits, for instance the bacterial culture studies with E. coli that generated nothing fundamentally new, mainly devolution. It certainly doesn't show an actual example of a complicated biological innovation arising by NDE processes. Surely the evolutionary biologists should be able to find an example of this in historical time with at least some single celled organism with huge populations and short generation times and strong new selective pressures due to human activities. Until then it's back to the "just so" stories again. magnan

"When I called you on your pitiful lies you changed your story" Darwinists are going to say anything and everything in an attempt to scuttle ID. dougcampo

JAM

The specific substitutions don’t matter for his calculation. The problem is that he claims that two are required, and that they have to occur simultaneously. It’s an exponential error. Behe claims that two amino acid residues must be changed in a single mutational event. There is no evidence to support his claim.

Go back and read your comment at #25. Your whole point was that Behe was proven wrong because some strains of CQR malaria plasmodium do not have a mutation at the 220 position. Now suddenly the problem is Behe wants two simultaneous mutations and the specific substitutions don’t matter? You claimed to give two cites where Behe said CQR required mutations at 76 and 220. The quotes clearly say no such thing so now you are changing your argument! Oh, by the way, your new argument is just as much BS as your previous one. Behe never claims the mutations must be in a single mutational event! What Behe requires is two mutations before CQR emerges, or at least CQR with a selective advantage. If CQR occurs after one mutation, say at 76, it doesn’t confer a significant selective advantage. Compare to atovaquone where in a clinical trial one in three people had a recrudescence of infection. Resistance to atovaquone is known to arise after a single mutation to the cytochrome b gene.

I’d like you to answer this question: why, given the availability of primary human cell cultures, did Behe choose to extrapolate from an organism whose mutation rate is impossible to directly measure, and which is very distantly related to us?

For one thing Behe isn’t studying human evolution, he is studying all of evolution. Behe is looking at malaria because of the large population sizes and the decades of observation.

White uses the verbs “developing” and “arising,” not “mutating” to describe what happens in populations. White uses the term “mutation rates” to describe what is measured more directly. White also uses “apparent” to explain that the rate Behe is peddling as reflecting the mutation rate is not real, and explains additional reasons why CQR alleles will have trouble becoming established in a population:

Wrong again. Here is what White says, “ This suggests that the per-parasite probability of developing resistance de novo is on the order of 1 in 10^20 parasite multiplications ." Notice JAM that is “per-parasite probability.” White distinguishes between “developing resistance” and the “de novo selection of resistance.” The calculation of the per parasite probability of developing resistance 10^20 is based only on the population of malaria in people who are sick, where probability of selection of de novo resistance is very high. As White states, “Taken together, the balance of evidence strongly favors acute symptomatic infection as the source of de novo antimalarial resistance.” Gametocytes carrying the resistance genes reach transmissible densities when the resistant biomass has expanded to a population size close to that necessary to produce illness (>107 parasites). Furthermore the relative fitness of CQR strains is .76 or .85, so it is not immediately deselected for lack of fitness in an environment without selective pressure for CQR. At any rate, Behe offers 10^20 as a rough estimate only so your is completely de minimas. You started off accusing Behe of claiming that CQR always required mutations at 76 and 220. When it was pointed out that Behe said “almost always” you claimed Behe contradicted himself. When Sal asked you to produce a quote where Behe contradicted himself you dishonestly cited pages that said nothing of the sort. When I called you on your pitiful lies you changed your story Jehu

Discovery Institute to fund...maybe a few million a year?

I wish that could be the case...

Discovery Institute’s Center for Science and Culture (its program on intelligent design and evolution) only spent $1.2 million in 2003, the year that Olson uses for his film. In 2004 it spent the same, and in 2005 it spent $1.6 million. Indeed, the budget for the entire Discovery Institute, including expenditures on non-intelligent design programs on transportation, technology, and other topics, has never reached $5 million. In 2003, the Institute as a whole spent $2.5 million, or half the figure cited by Olson. In 2004, it spent $3.5 million, and in 2005 it spent $3.9 million. These facts are publicly available for anyone to check on the Institute’s Form 990s posted at www.guidestar.com.

Patrick

This a great quote from David Berlinski that I want to share with you guys. ------------------------------------ The ID movement in its attack on Darwinism has simply articulated what many people instinctively feel. Darwin’s theory is plain nuts. It is not supported by the evidence; it has no organizing principles; it is incoherent on its face; it flies against all common experience, and it is poisonous in its implications. And another thing. It is easy to understand. Anyone can become an evolutionary biologist in an afternoon. Just read a book. Most of them are half illustrations anyway. It’s not like studying mathematics or physics, lot of head splitting stuff there. dougcampo

'I imagine they could fund a lab, a few researchers and a few graduate students for the several years the work would take out of their budget' This is a good idea Patrick. dougcampo

Jerry, I was not feigning concern for ID. Actually, if you thought that then obviously I failed to successfully get across what I was trying to say. dougcampo

In light of the above illustration, note how readily JAM will be willing to answer the question I posed to him. He obfuscated when I posed the question to him earlier. He's invited to answer my rephrased question. scordova

Let me suggest to the reader that some of the reading of Behe by the critics is uncharitable and highly disingenuous. To illustrate my point, let's say I had a math book and read on one page: f(x) = 2 and then on another page f(x) = x + 7 Would I go around arguing the author contradicted himself, or that the author is mistaken to say f(x) = 2, when he later says f(x) = x + 7? Of course not! However, the critics are pulling a similar stunt, and being very disingenuous by pulling Behe's 2 mutation quotes and computations out of context. JAM said Behe had a false claim on page 59, but what is the context. The context begins on page 57:

Suppose that P. falciparum need several separate mutations just to deal with one antimalarial drug. Suppose that changing one amino acid wasn't enough. Suppose that two different amino acids had to be changed before a beneficial effect for the parasite showed up. In that case...

Behe is creating an illustration to help us understand later discussions. Let me point the reader to how Michael Lynch pulled a similar unwholsome trick. See: Becoming a Jedi Master in the online ID Wars I gave the following illustration. ===== consider the following hypothetical example where Joe Proponent tries to advance a new theory in the field of medicine,

Joe Proponent writes: The research in our paper will focus on males who weigh 200 pounds and up…given that the focus of our research is on males who 200 pounds and up, males weighing 200 pounds and up will be the starting assumption in our model…there are other areas of research that can be pursued, namely males weighing less than 200 pounds, but that was not the focus of our research…..

and then StrawMan Critic comes along and tries to discredit Joe Proponent’s work, StrawMan Critic responds:

It is an unwarranted assumption that males weigh 200 pounds and up, we have numerous examples where this isn’t true. This is gross mischaracterization of the well established scientific literature. Here is a daunting list of peer-reviewed papers demonstrating that their assumptions that males weigh 200 pounds or more is totally unwarranted…therefore their model is false…therefore the conclusions of the research should be categorically rejected.

I hope the reader sees the flaws in StrawMan Crtitic’s argument. In the ID wars, such StrawMan arguments are very subtle, couched in technical language, and window dressed with citations to peer-reviewed literature. The screeds by the critics follow the same unwholose rhetorical form. The misrepresentation of the 2 mutation assumption is being promoted by the anti-ID critics in the same manner. scordova

dougcampo, Patrick Caldon provied no refutal of ID and neither has JAM. Patrick Caldon provided one interesting mutation in a dog and a couple mentions to some others and JAM is desperately trying to find some minutae wrong with Behe's book. If anything, the triviality of their replies is indicative that the basic premise of ID is even stronger. So I do not understand your feigned concern for the ID movement when as we often say, "Is this all they can come up with." jerry

dougcampo, You can answer the question I posed to JAM. If you choose not to, I consider your empty boast just that, an empty boast. If you don't answer, consider yourself univited to this thread as well. Sal scordova

JAM: I was away for a bit, otherwise I would have weighed in earlier:

I asked you: Simultaneoulsy for what? Simultaneously for any CQR to exist at all or just for that particular form of resistance?

I asked a simple question, and you obfuscated and didn't answer. The particular form of restance in question was the 76 and 220 resistance. You started obfuscating about dosage levels of CQ. That was just obfuscation. If a malarial strain has a 76 and 220 form of resitance in place, then it has a 76 and 220 form of resitance. So let me try to rephrase the question. Did Behe argue that a malarial strain will have CQ resistance if and only if it has the 76 and 220 mutations that confer resistance? You have 3 options: 1. yes 2. no 3. JAM doesn't know how to answer If you answer #3, then explain why you don't know how to answer. Answer, or consider yourself univited to this thread. scordova

dougcampo, Speaking as someone who's more theistic evolutionist than IDist, I don't think - even if Behe's claims turn out to be incorrect - the ID movement is in trouble. No more than evolution was in real trouble when Gould and Dawkins squared off, or the usefulness of junk DNA was (and is) pointed out. If a claim is proven to be wrong, the claim is either amended as needed, or discarded. That's science. At worst for ID is the realization that finding a way to detect design in biological processes (for example) is not available. In which case the argument places a foot more firmly in the realm of the philosophical rather than the scientific. Still, I think the ID movement has already succeeded in doing one thing - getting people to look at science with fresh eyes and attitudes, rather than the ones the high priests of science demand they rely on. nullasalus

dougcampo, I'd hardly describe my mutterings as overwhelming opposition, but I agree that it does seem a great shame that Michael Behe did not take the opportunity to do some experiments and properly characterize the development of CQR in its myriad forms. Numbers of people have done in-vitro work on this, the protocols seem reasonably well understood, what's needed now is a lot of grunt work to actually nail down what the probabilities of various sequences of mutation are. If it happened to contribute something to the ID debate, all for the good, but a precise description of the probabilities of CQR would almost certainly be invaluable in the construction of anti-malarial strategies, and could quite possibly save many hundreds of thousands of lives. This would surely be an excellent piece of work for the Discovery Institute to fund, I imagine they could fund a lab, a few researchers and a few graduate students for the several years the work would take out of their budget, maybe a few million a year? And if indeed the evidence (in this particular regard) did not end up showing the truth of ID, at least they would have the consolation that they had benefited the lot of humanity. Patrick Caldon

If ID people don't find something to discredit Darwinism soon you are going to lose the American people. I'm sorry, I want to believe and I really support everything you guys are doing but you are against what seems like overwhelming opposition. Both from the Darwinists and a culture that just doesn't care anymore. I hope you can develop some type of answer to what JAM and Patrick Caldon are saying. If not, then perhaps they are right and ID is just all wishful thinking. dougcampo

Patrick Caldon, Thank you for your comment. I will look into it and the other examples you mention. jerry

Well it looks as if JAM has completely discredited Behe's argument. dougcampo

Why should these rare successes produce so much variety in the alleles of mammals especially humans.

Humans are probably the least polymorphic mammals if one factors in reproductive success.

NS and genetic drift drive the elimination or reduction in the variety of alleles.

Then how do you explain the amount of current polymorphism? Doesn't selection drive polymorphism in histocompatibility loci, for example?

Mutations that produce changes in the allele structure are relatively rare,...

I don't know what you mean by "allele structure." The two words don't go together in any meaningful way. -----

Jehu: At first I thought you were being incredibly disingenuous but perhaps you have only failed to logically follow the argument. Let me help you.

Thanks.

Since your comment at #25 you have taken the erroneous position that Behe claimed that Chloroquine resistance (CQR) requires mutations at K76T and A220S.

The specific substitutions don't matter for his calculation. The problem is that he claims that two are required, and that they have to occur simultaneously. It's an exponential error.

...Here are the quotes for the whole world to see. First from page 59: “Let’s compare the two numbers for the odds of achieving resistance to atovaquone, where just one mutation is needed, versus chloroquine, where (presumably – since if a single mutation could help, chloroquine resistance would originate much more frequently) two are needed. ”

His presumption is incorrect, because clinical studies don't look at origination (mutation). They can only detect mutations after they have been filtered through selection.

And from page 60-61: “The likelihood that Homo sapiens achieved any single mutation of the kind required for malaria to become resistant to chloroquine–not the easiest mutation, to be sure, but still only a shift of two amino acids...” Do you see anything in either of those quotes about CQR requiring mutations at K76T and A220S? No. Because Behe never makes the claim.

Behe claims that two amino acid residues must be changed in a single mutational event. There is no evidence to support his claim.

Behe’s position is that CQR requires at least two mutations.

...in one event, but there's not a bit of evidence to support that.

Although mutations at K76T and A220S are “almost always present” in CQR strains, Behe does not base his position of CQR requiring at least two mutations on the need for mutations at K76T and A220S but on the fact that CQR resistance arises so rarely.

But since we are limited to looking after many rounds of selection both for and against the mutant, we aren't looking at the frequency of mutational events at all. With all the experimental systems available, why did Behe choose one that lacks cell culture methods? Why not look at mutation rates in cultured human cells instead?

How rarely does CQR arise? Here is a relevant quote from page 59.

De novo by mutation? We don't know. Behe doesn't know.

Spontaneous resistance to atovaquone [just one mutation needed] can be found in roughly every third sick person. Spontaneous resistance to chloroquine can be found perhaps in every billionth sick person …”

There's a big difference between what happens and what we can find long after it has been filtered by selection.

The research backing up Behe’s position is as follows:

He'd have to look directly at mutation. There are plenty of systems (including human ones) in which he could have done that, but he chose to base the probability of human evolution on something that is ridiculously distantly reltated to us and in which we lack the ability to measure mutation rates.

“Resistance to chloroquine in P. falciparum has arisen spontaneously less than ten times in the past fifty years (14). This suggests that the per-parasite probability of developing resistance de novo is on the order of 1 in 10^20 parasite multiplications.

What happens to any CQR mutants that originate in the mosquito phases, with no CQ around? They vanish. This is why White uses the terms "arisen" and "developing." It's not about mutation rates.

The single point mutations in the gene encoding cytochrome b (cytB), which confer atovaquone resistance, or in the gene encoding dihydrofolate reductase (dhfr), which confer pyrimethamine resistance, have a per-parasite probability of arising de novo of approximately 1 in 10^12 parasite multiplications (5).” N.J. White, Antimalarial Drug Resistance, J. Clin. Invest. 113:1084-1092 (2004).

So, let's read on from that point, shall we, and see who is doing the literature bluffing? But in the laboratory, much higher mutation rates than 1 in every 10E12 are recorded (12). What does that mean? Mutations may be associated with fitness disadvantages (i.e., in the absence of the drug they are less fit and multiply less well than their drug-sensitive counterparts). What does that mean in terms of extrapolating from appearance of resistance in a population to mutation rates? Another factor that may explain the discrepancy between in vitro and much lower apparent in vivo rates of spontaneous mutation is host immunity. Why does the author Behe cites (again, a review, demonstrating laziness) describe the ecological rates as "apparent," while Behe pretends that they are real?

Try to understand this JAM. Behe does not claim CQR requires mutations at K76T and A220S. Behe claims CQR requires two mutations. Behe bases this on the rarity of spontaneous CQR not on the explicit need for mutations at K76T and A220S.

I understand that. I also understand that there is no evidence supporting the claim that it requires two mutations. I also understand, as does the author of the paper Behe cites, that the rates of emergence of resistance in a population don't accurately reflect the mutation rate.

I would like to see one of Behe’s critics answer this question; if CQR can be achieved by single cumulative mutations, why has it arisen less than ten times in past fifty years?

Because the selection is intermittent at best, and reversed during each Plasmodium life cycle. I already noted this. Did you miss it? I'd like you to answer this question: why, given the availability of primary human cell cultures, did Behe choose to extrapolate from an organism whose mutation rate is impossible to directly measure, and which is very distantly related to us?

Art2: I’m wondering - is Behe using the review by White to argue that the frequency of occurrence of a double mutation is 1 in 10^20? Am I reading all of this - the discussion here, as well as EoE - correctly?

Apparently, despite the fact that White clearly understands that the emergence of CQR in a population is not a reflection of mutation frequency.

White derives the number by dividing the number of times de novo resistance has occurred by the population of the plasmodium under selective pressure of chloroquine resistance.

But White clearly realizes that this does not reflect the frequency of mutation.

It is a very simple and straight forward scientific observation that only those with a deep emotional commitment to Darwinism have trouble grasping.

White uses the verbs "developing" and "arising," not "mutating" to describe what happens in populations. White uses the term "mutation rates" to describe what is measured more directly. White also uses "apparent" to explain that the rate Behe is peddling as reflecting the mutation rate is not real, and explains additional reasons why CQR alleles will have trouble becoming established in a population: It is likely that this specific immune response directed against the immunodominant surface antigens will reduce the probability of the usually single mutant parasite ever multiplying sufficiently to transmit as for P. falciparum; there is only a 2–3% chance that the genetic event causing resistance would arise in the antigenically variant subpopulation that will expand to reach transmissible densities. Assuming an equal distribution of probabilities of spontaneous occurrence throughout the malaria parasites’ life cycle, the genetic event resulting in resistance is likely to take place in only a single parasite at the peak of infection. These genetic events may result in moderate changes in drug susceptibility, such that the drug still remains effective (e.g., the serine-to-asparagine mutation at position 108 in Pfdhfr that confers pyrimethamine resistance),... In the rare but important infection in which resistance arises de novo, killing of the transmissible sexual stages (gametocytes) during the primary infection does not affect resistance, because these gametocytes derive from drug-sensitive parasites. Gametocytes carrying the resistance genes will not reach transmissible densities until the resistant biomass has expanded to a population size close to that necessary to produce illness (>10E7 parasites) (38). Even if the resistant mutant does survive the initial drug treatment and multiplies, the chance that this will result in sufficient gametocytes for transmission is reduced as a result of both asexual-stage immunity (which reduces the multiplication rate and lowers the density at which the infection is controlled) and specific antigametocyte (transmission-blocking) immunity. Furthermore, other parasite genotypes are likely to be present, since infections are acquired continuously. These compete with the resistant parasites for red cells and increase the possibility of outbreeding of multigenic-resistance mechanisms or competition in the feeding anopheline mosquito. I highly recommend reading the White review; it is available online. There is nothing in it to suggest that the frequency of emergence of CQR Plasmodium represents the actual mutation rate, as Behe pretends it does. JAM

Hi Jehu, Is there another link? The one you gave to the Drug Resistance Paper isn't working for me. Charlie

Jehu, here's actually what White said: ""Chloroquine resistance in P. falciparum may be multigenic and is initially conferred by mutations in a gene encoding a transporter (PfCRT) (13). In the presence of PfCRT mutations, mutations in a second transporter (PfMDR1) modulate the level of resistance in vitro, but the role of PfMDR1 mutations in determining the therapeutic response following chloroquine treatment remains unclear (13). At least one other as-yet unidentified gene is thought to be involved. Resistance to chloroquine in P. falciparum has arisen spontaneously less than ten times in the past fifty years (14). This suggests that the per-parasite probability of developing resistance de novo is on the order of 1 in 10^20 parasite multiplications. The single point mutations in the gene encoding cytochrome b (cytB), which confer atovaquone resistance, or in the gene encoding dihydrofolate reductase (dhfr), which confer pyrimethamine resistance, have a per-parasite probability of arising de novo of approximately 1 in 10^12 parasite multiplications (5). To put this in context, an adult with approximately 2% parasitemia has 10^12 parasites in his or her body. But in the laboratory, much higher mutation rates thane 1 in every 10^12 are recorded (12)." It sounds to me like White was estimating the frequency of occurrence of "productive" mutations in, not one, but three different genes (PfCRT, PfMDR1, and an unidentified gene). At the very least, this passage is very, very ambiguous. I'm not sure it means what everyone here is assuming. Art2

Art2

I’m wondering - is Behe using the review by White to argue that the frequency of occurrence of a double mutation is 1 in 10^20? Am I reading all of this - the discussion here, as well as EoE - correctly?

White derives the number by dividing the number of times de novo resistance has occurred by the population of the plasmodium under selective pressure of chloroquine resistance. It is a very simple and straight forward scientific observation that only those with a deep emotional commitment to Darwinism have trouble grasping. Jehu

I'm wondering - is Behe using the review by White to argue that the frequency of occurrence of a double mutation is 1 in 10^20? Am I reading all of this - the discussion here, as well as EoE - correctly? Art2

jerry, That's simply not true. There have been numbers of lab experiments which have (in a lab environment) created new alleles. Likewise there are numbers of "field" experiments where the only sensible conclusion is that a new allele has been created. For instance, a mutation in the MSTN gene was recently discovered to produce a strain of fast whippets. This is a new allele which does not exist in wolves or other dog breeds, but makes whippets who carry it run faster. It has arisen naturally via a deletion event, and been selected for by breeders who like fast dogs. New alleles have been observed and sequenced in lab environments in numbers of species, including mice, S. cerevisiae, and brassica (cabbages) being the ones I've read papers about in the last several months. Patrick Caldon

I would like to see one of Behe’s critics answer this question; if CQR can be achieved by single cumulative mutations, why has it arisen less than ten times in past fifty years? Jehu

JAM, At first I thought you were being incredibly disingenuous but perhaps you have only failed to logically follow the argument. Let me help you. Since your comment at #25 you have taken the erroneous position that Behe claimed that Chloroquine resistance (CQR) requires mutations at K76T and A220S. You have maintained this position even after I showed you that Behe’s statement was not that that CQR requires mutations at K76T and A220S but that mutations at K76T and A220S were “almost always present” in CQR strains of malaria. Sal asked you to produce a quote showing where Behe stated that CQR required mutations at K76T and A220S. In response, at comment #44, you surprisingly claimed to have already posted a quote and cited to pages 59 and 60-61 of The Edge of Evolution. Neither of the Behe quotes that you cite say that CQR requires mutations at K76T and A220S. Here are the quotes for the whole world to see. First from page 59: “Let’s compare the two numbers for the odds of achieving resistance to atovaquone, where just one mutation is needed, versus chloroquine, where (presumably – since if a single mutation could help, chloroquine resistance would originate much more frequently) two are needed. ” And from page 60-61: “The likelihood that Homo sapiens achieved any single mutation of the kind required for malaria to become resistant to chloroquine–not the easiest mutation, to be sure, but still only a shift of two amino acids–the likelihood that such a mutation could arise just once in the entire course of the human lineage in the past ten million years, is minuscule–of the same order as, say, the likelihood of you personally winning the Powerball lottery by buying a single ticket.” Do you see anything in either of those quotes about CQR requiring mutations at K76T and A220S? No. Because Behe never makes the claim. Behe’s position is that CQR requires at least two mutations. Although mutations at K76T and A220S are “almost always present” in CQR strains, Behe does not base his position of CQR requiring at least two mutations on the need for mutations at K76T and A220S but on the fact that CQR resistance arises so rarely. How rarely does CQR arise? Here is a relevant quote from page 59. “ Spontaneous resistance to atovaquone [just one mutation needed] can be found in roughly every third sick person. Spontaneous resistance to chloroquine can be found perhaps in every billionth sick person …” The research backing up Behe’s position is as follows: “Resistance to chloroquine in P. falciparum has arisen spontaneously less than ten times in the past fifty years (14). This suggests that the per-parasite probability of developing resistance de novo is on the order of 1 in 10^20 parasite multiplications. The single point mutations in the gene encoding cytochrome b (cytB), which confer atovaquone resistance, or in the gene encoding dihydrofolate reductase (dhfr), which confer pyrimethamine resistance, have a per-parasite probability of arising de novo of approximately 1 in 10^12 parasite multiplications (5).” N.J. White, Antimalarial Drug Resistance, J. Clin. Invest. 113:1084-1092 (2004). Try to understand this JAM. Behe does not claim CQR requires mutations at K76T and A220S. Behe claims CQR requires two mutations. Behe bases this on the rarity of spontaneous CQR not on the explicit need for mutations at K76T and A220S. Jehu

Jam, Maybe you do not understand the process of Neo Darwinism. It eliminates not creates alleles. So the comments "So why would you expect the fewer number of reproductive events to produce all the variety in mammals."" Jam says "Because evolution predicts mostly dead ends (extinctions). When you look at all the variety in mammals, you’re only looking at the rare successes." I am afraid I don't understand your comment or how it answers my objection. Why should these rare successes produce so much variety in the alleles of mammals especially humans. NS and genetic drift drive the elimination or reduction in the variety of alleles. Mutations that produce changes in the allele structure are relatively rare, that is the theme of Behe's book so where did all the different types of alleles come from? This is one of the fundamental contradictions of NDE. jerry

JAM, is this what you're fixated on? "When chloroquine is no longer used to treat malaria patients in a region, the mutant strain of P. falciparum declines and the original strain makes a comeback, indicating that the mutant is weaker than the original strain in the absence of the toxic chloroquine." pp. 50-51 PaV

No, why? JAM

What about all the variety in dog breeds in a short time frame?

Been reading Dawkins, lately? ;) Patrick

Sal: What other claim JAM?

The one I directly quoted, Sal, that spans the break between pages 60 and 61.

Unless you back it up with specific citations you are uninvited to this thread. You’re invited to discuss, you’re not invited to fabricate or insinuate things Behe didn’t say.

I'm not fabricating anything. Behe assumes that two substitutions are required twice on p. 59 as well.

Simultaneoulsy for what? Simultaneously for any CQR to exist at all or just for that particular form of resistance?

For me to answer, you'd have to explain what you mean by "form." CQR is simply an empirical measurement that varies by resistant dose. Or are you using "form" to mean "CQR-resistant allele"?

Behe pointed out that that there could be other forms of resistance. However, he was pointing out why this particularly effective one took so long to happen.

No, he gave the reason why (the relative fitness of mutant and wild-type reverses in the absence of CQ) on pages 50-51 but ignored it in his conclusion. What he doesn't mention, and you need to think about, is that this flip-flop of selection pressure goes on in the mosquito in every infectious cycle, so that the mutants are at a disadvantage. We're only looking at the products in populations after selection, not at the raw production of variation.

Don’t parade the strawman that he excluded other possibilities a priori.

I'm not. He has an interesting technique of mentioning things he later ignores in his calculations and conclusions.

Let me give you one from page 61: Someone might object that, since there are thousands of other proteins in an organism, much other DNA, and many other kinds of mutations than just amino acid changes, aren’t the odds of finding some beneficial complex of mutations much better than the odds of finding just the specific complex of mutations we isolated?

That someone wouldn't be me, and I make my living designing mutations that enable proteins to utilize new substrates that aren't found in nature.

The dual PfCRTA144T and PfCRTL160Y mutations are more remote than a single PfCRT220 mutation.

Yes, but since either of them get us to resistance, we add their probabilities together.

It is not surprising that the 220 arose apparently more frequently in several independent places than the A144T and L160Y.

Since we don't know exactly what any of the substitutions do, we can't look at the products of selection as populations and pretend that we are looking directly at the products of mutation.

Baloney. That’s exactly the point of why it took so long for CQR to emerge.

No, the emergence is the product of selection on heterogeneous populations. Behe explicitly gives the reason on p. 50-51, but then abandons it in his conclusion.

There were not many cumulative paths, because had their been, the resistance would have happened in a flash!

The resistance is a product of intermittent, frequently reversed selection. Remember, in each cycle, the Plasmodium is in a mosquito in the absence of CQ, so the wild-type is favored over the mutant.

You apparently don’t get the logic of the argument.

Sal, I get the logic. My problem is with the assumptions. You are assuming that Plasmodium is asexually reproducing under constant selection favoring the resistant allele, as in the case of E. coli in the presence of an antibiotic. The reality is that Plasmodium is sexually reproducing under selective pressures that literally reverse every couple of weeks. Behe notes that the selective pressure reverses in people without CQ, but apparently doesn't realize that this is the case in mosquitoes, too.

The requirement for a long time for resistance to CQ implies there are few effective cumulative paths.

No, it's a product of reversing fitness landscapes.

You’re equivocating the sense of “we assume that two specific substitutions must occur simultaneously”. There are two senses of this phrase: 1. these two are simultaneously needed for a particular kind of CQR (namely a 76 and 220 form of resistance) 2. these two are simultaneously needed for ANY CQR. You are arguing #2 and Behe is arguing #1.

Which is he assuming in his calculation?

Since Behe is not arguing for a single route to CQR,...

Isn't that implicit in his calculation of 10E-20? If you disagree, please walk me through the calculation.

You fail to grasp that the length of time involved in forming CQR is a measure of the amount of cumulative pathways avaialable.

Sal, the length of time is a measure of the strength of selection on populations. Behe even notes that in the absence of CQ, the wild-type has an advantage over the CQR mutants.

If you assert there are lots of pathways that were accessible, you’ll have to account for why it didn’t happen quickly as opposed to decades.

I'm not asserting that. The constantly reversing selection in each reproductive cycle accounts for the speed.

No it is not, for the reasons Behe outlined. Were that path readily available it would have been found far faster. You fail to grasp the logic.

Again, it's about the underlying assumptions. When you genotype clinical isolates of Plasmodium, you are looking at the products of a mixture of wild-type and mutant plasmodia subjected to a constantly-changing (and even reversing, as Behe himself notes) fitness landscape. JAM

jerry: Thus, the relative lack of response to the sickle cell mutation or other means of fighting the parasite has not produced much. So why would you expect the fewer number of reproductive events to produce all the variety in mammals.

Because evolution predicts mostly dead ends (extinctions). When you look at all the variety in mammals, you're only looking at the rare successes. What about all the variety in dog breeds in a short time frame?

Behe has put into numbers what we all instinctively sense from looking at the evidence. So the quibbling over minor changes in amino acid sites may be another form of bluffing and distracting from the real issues.

Behe made this calculation the foundation of his thesis, so pointing out problems with it is neither quibbling nor bluffing. If they are insignificant, then it seems that you should blast Behe for basing his argument on this estimate. There are many additional real issues, for example Behe's neglect of far more rapid mechanisms of genetic variation, such as recombination. The Plasmodium calculation illustrates that issue, too, since Behe discounts the much more likely model of two independent substitutions being put together by recombination. His calculation assumes asexual reproduction and ignores sexual reproduction. JAM

Thanks for the clarification. With regard to the sequential argument, it is helpful seeing the probability there since the wedge argument here is that CQR is "possible" through sequential changes. Seeing that it doesn't functionally change the underlying probability is helpful, as is the clarification of types of CQR Behe is arguing. rswood

Sal, It wasn't a correction. It was a further clarification, which all of us would do well to keep in mind, since this shoots down the usual arguments from those defending the Modern Synthesis. PaV

Thanks PaV for the correction. Sal scordova

"1. these two are simultaneously needed for a particular kind of CQR (namely a 76 and 220 form of resistance)" Would it better to say: 1. these two need to both be present for a particular kind of CQR (namely a 76 and 220 form of resistance) There's confusion here, I think, between "simultaneous" and sequential. If the odds of just "one" needed mutation is 1 in 10^12, and the odds of getting both mutations in one cell at the very same time is 10^20, well, I guess it's not impossible for 10^20 malarial cells to be produced without forming any mutation, and then, all of a sudden, to mutate at both positions in the very same reproductive cycle, but it is much more probable---and experimentally verified---that after 10^12 malarial cells have been produced that one of these cells has "one" of the two needed mutations. Then it is simply a matter of waiting for the second to occur from that cell, which has the odds of 1 in 10^8 (the genome size of Plasmodium). This works out to 1 in 10^20, and, lo and behold, Behe uses someone else's calculations to demonstrate the validity of his thinking: Nicholas White of Mahidol University in Thailand points out that if you muliply the number of parasites ina person who is very ill with malaria times the number of people who get malaria per year times the number of years since the introduction of chloroquine, then you can estimate that the odds of a parasite developing resistance to chloroquine is roughly one in a hundre billion billion. In shorthand scientific notation, that’s one in 10^20. Continuing on, it doesn't matter which of the two mutations occurs first,(with the odds of 1 in 10^12 of occuring---based on studies), but the second will occur with the odds of it occurring being 1 in 10^8 (genome size of Plasmodium), giving a total probability of 1 in 10^20. These are "sequential" mutations; not "simultaneous" mutations. But the cumulative odds are the same; i.e., it doesn't matter if they happen at the same time, or if first there is one, and then the other. But, statistically, one would SUPPOSE that they happened sequentially. PaV

JAM wrote: Yes, that demonstrates that he knows that his hypothetical claim elsewhere in the book is false:

What other claim JAM? Unless you back it up with specific citations you are uninvited to this thread. You're invited to discuss, you're not invited to fabricate or insinuate things Behe didn't say.

JAM wrote: Have you found any data to support Behe’s assertion that both substitutions must occur simultaneously?

Simultaneoulsy for what? Simultaneously for any CQR to exist at all or just for that particular form of resistance? Behe pointed out that that there could be other forms of resistance. However, he was pointing out why this particularly effective one took so long to happen. Don't parade the strawman that he excluded other possibilities a priori. Let me give you one from page 61:

Someone might object that, since there are thousands of other proteins in an organism, much other DNA, and many other kinds of mutations than just amino acid changes, aren't the odds of finding some beneficial complex of mutations much better than the odds of finding just the specific complex of mutations we isolated? No. Many, many other mutations in addition to the ones we discusses have popped up by chance in the vast worldwide malarial pool over the course of a few years. In fact mutations in all of the amino acid position in all of the proteins of malaria--taken both one and two at a time--can be expected to occur by chance during the same stretch of time. And other types of mutations besides just changes in amino acids would also occur (such as insertions, deletions, inversions, gene duplications, mobile DNA transpositions, changes in regulatory regions, and others, perhaps even including whole genome duplication---some of these types of mutations are described in the next chapter). Although some other mutations in some other proteins are thought to contribute to chloroquine resistance, none are nearly as effective as that in PfCRT. That means that of all the possible mutations in all of the different proteins of malaria, only a miniuscule number have the ability to help at all against chlorquine, and only one, PfCRT, is really effective.

The dual PfCRTA144T and PfCRTL160Y mutations are more remote than a single PfCRT220 mutation. It is not surprising that the 220 arose apparently more frequently in several independent places than the A144T and L160Y.

JAM asked: Sal, even if we assume that two specific substitutions must occur simultaneously without any evidence,

Baloney. That's exactly the point of why it took so long for CQR to emerge. There were not many cumulative paths, because had their been, the resistance would have happened in a flash! You apparently don't get the logic of the argument. The requirement for a long time for resistance to CQ implies there are few effective cumulative paths. You're equivocating the sense of "we assume that two specific substitutions must occur simultaneously". There are two senses of this phrase: 1. these two are simultaneously needed for a particular kind of CQR (namely a 76 and 220 form of resistance) 2. these two are simultaneously needed for ANY CQR. You are arguing #2 and Behe is arguing #1.

what does each alternative substitution (including those at the same codon) do to Behe’s probability estimate, even if we assume that they must occur simultaneously?

Since Behe is not arguing for a single route to CQR, it doesn't affect the probability of 76 and 220 resistance. Because, assuming we have something like 76 combined with A144T+L160Y, that is statistically more remote than the 76 and 220 pathway. Therefore appeals to this even more remote path as being substantially helpful in terms of accumulation is illogical. It is not surprising A144T+L160Y appear to be rarer than the single 220 substitution. The fact that A144T+L160Y took time to emerge is suggestive there are not very strong cumulative paths to this form of resistance either. You fail to grasp that the length of time involved in forming CQR is a measure of the amount of cumulative pathways avaialable. If you assert there are lots of pathways that were accessible, you'll have to account for why it didn't happen quickly as opposed to decades.

He [Miller} doesn’t have to insinuate anything. It’s a fact.

No it is not, for the reasons Behe outlined. Were that path readily available it would have been found far faster. You fail to grasp the logic. scordova

I botched the quoting at the end of my comment and repeated myself unnecessarily... He doesn’t have to insinuate anything. It’s a fact.

...therefore there is surely a cumulative,sequential, 1-amino-acid-at-at time route, the Miller has made yet another faux pas. Good for him!

Sal, even if we assume that two specific substitutions must occur simultaneously without any evidence, what does each alternative substitution (including those to different aa residues at the same codon) do to Behe’s probability estimate? JAM

One of the things to keep in mind as part of Behe's premise is that there have been more reproductive events in malaria in the last 1000 or 10,000 years then all the mammal reproductive events in history. Thus, the relative lack of response to the sickle cell mutation or other means of fighting the parasite has not produced much. So why would you expect the fewer number of reproductive events to produce all the variety in mammals. Behe has put into numbers what we all instinctively sense from looking at the evidence. So the quibbling over minor changes in amino acid sites may be another form of bluffing and distracting from the real issues. The real issue is that not much happens when things are left alone to evolve or are threatened with resistance (sickle cell) or annihilation (drugs.) jerry

Jehu: Behe’s argument is primarily about the probabilities of certain adaptations arising, not whether a mutation at A220S is is required for chloroquine resistance.

Behe's probability is explicitly derived from that assumption.

Furthermore, Behe never says that a mutation at 220 is always present in chloroquine resistant malaria. Here is what Behe writes in The Edge of Evolution “ However, the same two amino acid changes are almost always present. – one switch at position number 76 and another at position 220. ”

Yes, that demonstrates that he knows that his hypothetical claim elsewhere in the book is false: “The likelihood that Homo sapiens achieved any single mutation of the kind required for malaria to become resistant to chloroquine–not the easiest mutation, to be sure, but still only a shift of two amino acids...”

Notice the “almost.”

I definitely did. Do you notice that "almost" explicitly contradicts his claim that they are required?

The paper you bring up, where 220 is absent, shows that where there is no mutation at 220 there are two other mutations, A144T and L160Y. These two mutations as a substitute for 220 are less probable than a single mutation at 220 and less common as well. This findingcertainly does not damage Behe’s argument.

It makes his estimate of probability exponentially wrong, so it irreversibly damages his argument. Have you found any data to support Behe's assertion that both substitutions must occur simultaneously? Even if both were required, if that assertion is false, his probability is wrong by many orders of magnitude.

If you are going to try to claim that a single mutation can confer chloroquine resistance then please show some evidence of that and explain why chloroquine resistance so rarely arises independently.

I don't have to claim that a single mutation is sufficient. I only have to show that Behe's assumption that two specific substitutions must occur simultaneously is unsupported to show that Behe's estimate of probability is exponentially wrong.

Given the large populations of Malaria in a single infected human and the large number of humans infected, it seems like the probability of any singe amino acid substitution occurring in a given year is quite good.

First, there aren't large populations of Malaria in anyone, because malaria is a disease. The organism we are discussing is Plasmodium.

Therefore, if chloroquine resistance can be conferred by a single mutation, we should see more novel strains of resistant malaria than we do.

Sorry, but that doesn't cut it. You and Behe are omitting the far more rapid mechanisms of variation from the probability estimate, particularly recombination, and completely ignoring sampling bias. You need to realize that always (not almost always) observing both substitutions is absolutely necessary if Behe's hypothesis is correct, but in no way sufficient evidence to state it as fact, which Behe did. For example, the near-universal presence of at least two substitutions in clinical isolates can be explained by independent occurence of each, coupled with an exponential improvement in resistance for the double mutant. The facts of intense selection coupled with recombination then predict that it will be almost impossible to find a resistant isolate with only one substitution.

JAM: Would you please look at Table 1, and tell me if it shows that that resistant Plasmodium exist that lack either K76T or A220S? Jehu: I looked at the table and I don’t see where the table describes anything of the sort.

Every isolate that has "A" at the eighth position shows that.

The body of the article does talk about A220S being replaced by A144T and L160Y in that P2a and P2b strains but this has already been discussed.

How would that change Behe's probability calculation, and where is the evidence that they must occur simultaneously instead of sequentially? And where is the evidence that isolates with two substitutions were not derived from recombination events?

Sal: But I believe Jehu has shown your argument is a misrepresentation.

I don't think I've misrepresented anything. Jehu only pointed out that Behe contradicted himself in the book, which is pretty damaging to Behe's position, not Miller's.

As long as critics are making well-conceived objections (meaning objections that ID proponents themselves might entertain), then I welcome hearing them on my threads.

You might want to reconsider that definition.

Miller fails to inform us that resistance with mutaion 220 missing are equally challenging to evolve because instead of 1 mutation (for 220), to achieve similar effect, one needs 2 mutations in positions 144 and 160, which would suggest since we are dealing with 2 instead of 1, the situation is even more improbable!

Behe fails to inform us that both substitutions don't have to happen in the same Plasmodium, because of a widely-understood phenomenon called recombination. How does considering recombination numerically change Behe's probability estimate?

Thus it doesn’t help miller’s insinuation of a fast sequential path whatsoever.

Sure it does, especially when one considers recombination. Where is the evidence that the substitutions must occur simultaneously?

The point Behe was trying to illustrate in his book was why it took so long for resistance to emerge was that several of the resistance mechanisms require simultaneous changes to have a significantly selectable effect.

But Behe presented no evidence demonstrating that they must occur simultaneously.

And yes, Behe said, “almost”. For miller to insinuate Behe said “never” is a misrepresentation.

Behe's other statement, that they are required, means that one can never be found without the other. Even if they are, that is not sufficient evidence to claim that they must occur simultaneously.

Right on! If Miller is insinuating that because there are CQR (Chloroquine Resistant Malaria’s) like the Philippine strain that do not have te 220 mutation,

He doesn't have to insinuate anything. It's a fact. therefore there is surely a cumulative,sequential, 1-amino-acid-at-at time route, the Miller has made yet another faux pas. Good for him! Sal, even if we assume that two specific substitutions must occur simultaneously without any evidence, what does each alternative substitution (including those at the same codon) do to Behe's probability estimate, even if we assume that they must occur simultaneously? If you can't find any evidence that they must occur simultaneously, why are you trying to place the evidentiary burden on Miller instead of Behe? JAM

Jerry, Your comment was fine. I always look forward to hearing from you. Sal scordova

Jehu astutely observes: If you are going to try to claim that a single mutation can confer chloroquine resistance then please show some evidence of that and explain why chloroquine resistance so rarely arises independently. Given the large populations of Malaria in a single infected human and the large number of humans infected, it seems like the probability of any singe amino acid substitution occurring in a given year is quite good. Therefore, if chloroquine resistance can be conferred by a single mutation, we should see more novel strains of resistant malaria than we do.

Right on! If Miller is insinuating that because there are CQR (Chloroquine Resistant Malaria's) like the Philippine strain that do not have te 220 mutation, therefore there is surely a cumulative,sequential, 1-amino-acid-at-at time route, the Miller has made yet another faux pas. Good for him! scordova

JAM, By the way, welcome to the weblog. That was one of the more well-conceived objections that I've heard so far. But I believe Jehu has shown your argument is a misrepresentation. As long as critics are making well-conceived objections (meaning objections that ID proponents themselves might entertain), then I welcome hearing them on my threads. Miller fails to inform us that resistance with mutaion 220 missing are equally challenging to evolve because instead of 1 mutation (for 220), to achieve similar effect, one needs 2 mutations in positions 144 and 160, which would suggest since we are dealing with 2 instead of 1, the situation is even more improbable! Thus it doesn't help miller's insinuation of a fast sequential path whatsoever. The point Behe was trying to illustrate in his book was why it took so long for resistance to emerge was that several of the resistance mechanisms require simultaneous changes to have a significantly selectable effect. We don't know for sure how many possible routes to Chloroquine resistance there are, we know of a few, but evidentially speaking, there are not many, otherwise Chloroquine (CQ) resistance (R) from scratch would have developed much faster. And yes, Behe said, "almost". For miller to insinuate Behe said "never" is a misrepresentation. PS I bet when Behe gives his comprehensive response, he'll squash those misrepresentations by Miller. Right now Behe is merely throwing up his shield at amazon, the counter strike is yet to come. scordova

JAM

No, all it takes is one exception to show that Behe’s claim that both K76T and A220S are required is false.

Behe’s argument is primarily about the probabilities of certain adaptations arising, not whether a mutation at A220S is is required for chloroquine resistance. Furthermore, Behe never says that a mutation at 220 is always present in chloroquine resistant malaria. Here is what Behe writes in The Edge of Evolution “ However, the same two amino acid changes are almost always present. – one switch at position number 76 and another at position 220. " Notice the “almost.” The paper you bring up, where 220 is absent, shows that where there is no mutation at 220 there are two other mutations, A144T and L160Y. These two mutations as a substitute for 220 are less probable than a single mutation at 220 and less common as well. This findingcertainly does not damage Behe’s argument. If you are going to try to claim that a single mutation can confer chloroquine resistance then please show some evidence of that and explain why chloroquine resistance so rarely arises independently. Given the large populations of Malaria in a single infected human and the large number of humans infected, it seems like the probability of any singe amino acid substitution occurring in a given year is quite good. Therefore, if chloroquine resistance can be conferred by a single mutation, we should see more novel strains of resistant malaria than we do.

I think the problem is that you are looking for quotes, when the evidence that clearly contradicts Behe’s claim is in the data, specifically Table 1 of this paper (free PDF available): Antimicrob Agents Chemother. 2005, 49:2102-5._Origin and dissemination of chloroquine-resistant Plasmodium falciparum with mutant pfcrt alleles in the Philippines._Chen N, Wilson DW, Pasay C, Bell D, Martin LB, Kyle D, Cheng Q. I think that’s why Miller says that Behe waves off the evidence. Would you please look at Table 1, and tell me if it shows that that resistant Plasmodium exist that lack either K76T or A220S?

I looked at the table and I don’t see where the table describes anything of the sort. The body of the article does talk about A220S being replaced by A144T and L160Y in that P2a and P2b strains but this has already been discussed. Jehu

This is a little bit off target since it doesn't involve Miller, but does involve Behe and his new book. There was an article in the NY Times this week that is relevant to Behe's thesis. I am afraid to put the link in this site because it will send the post to the nether world. But go to google news and put in "lenski source:new_york_times" and the article will appear in the search. Behe lists Lenski's work as one of the main sources for his conclusions though you will not get that from the article. It is a fascinating article about the changes in bacteria in 20 years of breeding. It is NDE in action. However, it no ways refutes the basic objection to NDE that it works but only on trivial changes in micro organisms. However, some of the changes that did happen are extremely interestiing. There have been lots of mutations that have affected the organism's ability to thrive. This work should be a future addendum to Behe's book. When questioned by Jason Rosenhouse about future ID research, Behe specifically mentioned Lenski as something that should be carried on. Rosenhouse failed to understand how this would support the ID premise. Sorry Salvador that this is not about Miller but it has relevance to the overall discussions going on. Has Miller or anyone else cited Lenski's work in regards to Behe's book? jerry

Here was Behe calling Miller on some more distortions a while back: A True Acid Test: Response to Ken Miller scordova

Miller have just posted a summary of his review on Amazon.com. It would be nice if someone who can comment on reviews at amazon links to this post. IDist

"Ken Miller, the honest Darwinist"? To me it is more like "Ken Miller, the dishonest Darwinist." Ken Miller is a hypocrite. He was an expert witness for the plaintiffs in both the Kitzmiller v. Dover and Selman v. Cobb County evolution disclaimer cases. He says that his own belief in Darwinism is supported by his religious beliefs, yet he misuses the Constitution's establishment clause to try to suppress the views of critics of Darwinism whose criticisms are often not based by religious beliefs. And I was astonished that the plaintiffs in those two cases had the chutzpah to choose a theistic evolutionist as an expert witness in an establishment clause lawsuit. Larry Fafarman

What we need is a wikipedia for evolution, but an honest one

How about this one: ResearchID.org EndoplasmicMessenger

Sal: Did Miller forget this study? No, all it takes is one exception to show that Behe's claim that both K76T and A220S are required is false. or even in the paper Miller cites and accuses Behe of waving off: Does Miller accuse Behe of waving the paper off, or of waving off the evidence? mutations K76T and Ala220 Ser (A220S) appear to be the most reliable markers predicting CQ resistance. That's true, but it doesn't support Behe's claim that they are required, much less that they must occur simultaneously: "The likelihood that Homo sapiens achieved any single mutation of the kind required for malaria to become resistant to chloroquine--not the easiest mutation, to be sure, but still only a shift of two amino acids--the likelihood that such a mutation could arise just ?once in the entire course of the human lineage in the past ten million years, is minuscule--of the same order as, say, the likelihood of you personally winning the Powerball lottery by buying a single ticket." I think the problem is that you are looking for quotes, when the evidence that clearly contradicts Behe's claim is in the data, specifically Table 1 of this paper (free PDF available): Antimicrob Agents Chemother. 2005, 49:2102-5. Origin and dissemination of chloroquine-resistant Plasmodium falciparum with mutant pfcrt alleles in the Philippines. Chen N, Wilson DW, Pasay C, Bell D, Martin LB, Kyle D, Cheng Q. I think that's why Miller says that Behe waves off the evidence. Would you please look at Table 1, and tell me if it shows that that resistant Plasmodium exist that lack either K76T or A220S? and the final nail in Miller’s argument: ...they [76 and 220] rarely, if ever, occur alone That contradicts Behe right there. For them to be required, they necessarily must NEVER be found alone, not merely rarely. Even if that were true, that is a mere prediction; it would not be sufficient to conclude that they are both required or that they must occur simultaneously instead of sequentially. PS For the situation where 220 is missing but we still find CQR, see comment on A144T/L160Y But if the substitution at 220 is missing in any case at all, Behe's claim that it is required is false. It looks even worse for Behe because elsewhere in the book, he notes that they are "almost always" present. Almost isn't good enough. Scientific citation is generally about the evidence found in the figures and tables, not quoting. JAM

Regarding Super Monkey Ball evolution.

What I find interesting, is that this argument was not used by any of the major reviewers of Behe’s book, as far as I know. It is just popular among Darwin’s internet fans.

I agree that right now only the bloggers are advocating Super Monkey Ball evolution. However, in the past, Allen Orr has demonstrated that he is at heart a believer in Super Monkey Ball . Here is his quote, Consider fitness functions that are as unsmooth as you like, i.e., rugged ones, having lots of peaks and few long paths up high hills. (These are the best studied of all fitness landscapes.) Now drop many geographically separate populations on these landscapes and let them evolve independently. Each will quickly get stuck atop a nearby peak. You might think then that Dembski's right; we don't get much that's interesting. But now change the environment. This shifts the landscape's topography: a sequence's fitness isn't cast in stone but depends on the environment it finds itself in. Each population may now find it's no longer at the best sequence and so can evolve somewhat even if the new landscape is still rugged. Different populations will go to different sequences as they live in different environments. Now repeat this for 3.5 billion years. Will this process yield interesting products? Will we get different looking beasts, living different kinds of lives? My guess is yes. So Allen Orr's guess is that evolution does in fact play Super Monkey Ball. Of course, I am not aware if Orr has or will review Behe this time around, if he does we will have to see if he brings back Super Monkey Ball. Jehu

The problem with the Super Monkey Ball agrument is pretty obvious. But it is still a favorite of the Darwinian faithful.

What I find interesting, is that this argument was not used by any of the major reviewers of Behe's book, as far as I know. It is just popular among Darwin's internet fans IDist

Another common criticism of I see of Behe is that he does not account for dynamic fitness landscapes. This is the Super Monkey Ball argument. According to this argument, a dynamic fitness landscape helps guide a gene through sequential mutations to eventual complexity by preventing the gene from getting stuck at a local maximum. This works exactly like Super Monkey Ball where a tilting landscape guides a monkey in a ball through a maze to an eventual goal. The problem with the Super Monkey Ball agrument is pretty obvious. But it is still a favorite of the Darwinian faithful. Jehu

Miller writes,

Behe obtains his probabilities by considering each mutation as an independent event, ruling out any role for cumulative selection, and requiring evolution to achieve an exact, predetermined result.

This is a common theme criticizing Behe - Behe ignores the possibility of cumulative selection. Sean Carroll made the same claim. But is it true? No it is not. In The Edge of Evolution Behe spends some time discussing examples of cumulative selection. For example, Behe actually spends more than a full page discussing pyrimethamine resistance by cumulative selection. Here is quote, "Although the first mutation (at position 108 of the protein, as it happens) grants some resistance to the drug, the malaria is still vulnerable to larger doses. Adding more mutations (at positions 51, 59, and a few others) can increase the level of resistance." Explaining how he covered cumalative selection, Behe writes in his Amazon blog, "I discuss gradual evolution of antifreeze resistance, resistance to some insecticides by “tiny, incremental steps — amino acid by amino acid — leading from one biological level to another”, hemoglobin C-Harlem, and other examples, in order to make the critically important distinction between beneficial intermediate mutations and detrimental intermediate ones." So the ignoring cumulative selection argument is a complete strawman. Behe's position is that the creative power of cumulative selection is extremely limited and he backs up this position with real world examples of astronomical populations getting very limited results with it. This is something the critcis don't really address. Jehu

For the record, Miller may have been thinking of this when he discussed CQRs without the 220 mutation Origin and Dissemination of Chloroquine-Resistant Plasmodium falciparum with Mutant pfcrt Alleles in the Philippines :

Palawan, though separated by sea, shares a closely related dominant pfcrt allelic type with Central Luzon. The novel mutations in P2a and P2b alleles (A144T and L160Y) were not found in a wild-type pfcrt gene, indicating that these mutations in the CQR isolates were not inherited from chloroquine-susceptible parasites in the Philippines. In addition, none of the Philippine isolates with A144T and L160Y mutations (n = 48) carried the A220S mutation very commonly seen in CQR parasites elsewhere. The mutually exclusive presence of A144T/L160Y with A220S suggests that A144T and L160Y may play a similar role as A220S in CQR.

But it would be hard to argue that this supports Miller's point since instead of a mutation at a single position of A220S we are dealing with two! A144T/L160Y scordova

"I just wish that these journals give Behe the opportunity to respond, but this is almost impossible." Is that impossible? Isn't there some way to insist that Behe get a chance to respond to all the falsehoods and rhetoric being relayed in the numerous reviews of his book? Forthekids

Mung wrote: Sal, you should have been quoting me rather than Mike Gene, as I have priority on pointing out Miller’s blatant falsehoods.

I'd be happy to publish your colleciton on my website. Is there a way I can get a hold of it? Sal scordova

One of the problems is that there is no place where there is a clearinghouse for information on all the aspects of the debate, not just on Ken Miller per se.

It's in the works. I'll have a spot for "honest" Ken Miller and Richard "sweetness" Dawkins. scordova

Miller and his comdrades in arms all defend evolution with the basic mentality of "the ends justify the means". It's really that simple. It's a glaringly obvious ethical scandal to any and all people who see how they operate and know where and when they utilize hyperbole, lies, etc, as examples of truisms. I wonder if they know they have no integrity or if they have a psychological condition which causes them to see what they do as being honest. Either way it just shows how pathetically the so-called enlightened are in reality wallowing in the mire while shouting to one all that they are dancing on the ceiling to the tune of the truth the light and the way. They build their maze of lies in the hopes of destroying people's faith in their perceived enemy, in actual reality their real enemy is their arrogance and small mindedness. Their mad maze of lies only ends up keeping all who enter it hopelessly lost. mentok

The problem here is that you are preaching to the choir. You need to catagorize these errors/bluffs to persuade anyone. A website would be a great idea except it calls for time and might end with being blackballed by the scientific community. Any way, kudos to the op. bork

Mike1962, One of the problems is that there is no place where there is a clearinghouse for information on all the aspects of the debate, not just on Ken Miller per se. It is all in bits and pieces. The other side is better organized and better funded. I suggested the other day that a webpage be developed that would point out all the flaws on the COE postion paper on conflating creationism with ID. It would be easy, just compare what the COE said and what the truth is. I doubt anyone would do it but it would take about a day to do that particular project. How about hiring a summer intern at the Discovery Institute. Maybe when I retire, I will do something but I am sure that like me all of us are too busy with life to do this. What we need is a wikipedia for evolution, but an honest one. jerry

Miller is not interested in the science, he is only interested in making money by selling books to rubes. rrf

Is there a website or page that has a compendium of all the errors and misrepresentations of Miller and the rest to which I could point people who enquire? If not, it would be a good project for someone. mike1962

"Don’t be too hard on Ken. He has a lot invested in the darwinian paradigm" My guess is that Ken Miller has a large financial stake in the outcome. He makes a couple hundred thousand dollars a year off Darwinism. Others do too. It is a growth industry on both sides. Follow the Money! jerry

Don't be too hard on Ken. He has a lot invested in the darwinian paradigm and has based a lot of his theology on it I suspect, so like the atheists he has a definite religious stake in the outcome. If he is deliberately misrepresenting Mike then he is probably doing so because he sees him as a threat, but in reality he is probably not intentionlly doing it, it is just that the work of Mike's sets off alarm bells that threatens Ken's theology and this impairs his reading comprehension because he sees it all through a "this is a threat to my theology" lense, like many of the atheists who get so upset. So cut the guy some slack and give him the benifit of the doubt. You are asking a lot from him to change his mind on this issue and it may not be possible for him to. He is simply suffering from Mortons Demon. Jason Rennie

I still have to pick up Behe's book, though I wouldn't go so far as to question Miller's christianity. Admittedly, I'd love to see him turn his guns more on Dawkins and the like, who I see as fanning the 'evolution is atheism!' flames. I wish the people in these debates could discuss things in calmer, more relaxed ways. If they disagree with Behe, why be so smug an animated about it? Likewise for people who disagree with Miller. But, that's hard to manage nowadays. Either way, interesting to see this review's gotten such things wrong. nullasalus

I really like your posts Sal. But the problem is that no matter how many errors, distortions, misrepresentations, misquotes or misunderstandings of Behe's work in the reviews by the darwinist elite, the reviews will be endelessly cited as refuted Behe. After all, these are published in Nature and Science etc. In fact, Richard Dawkins in his pathetic review cites Coyne's as "devastating". I just wish that these journals give Behe the opportunity to respond, but this is almost impossible :( IDist

In this work he claimed that complex biochemical systems have a property he called "irreducible complexity". Irreducibly complex structures, such as the bacterial flagellum, could not have evolved because they lack any selectable function until all of their component parts are in place. As he wrote, "any precursor to an irreducibly complex system is by definition nonfunctional", since every part of such a system had to be in place for natural selection to favour it. Ken Miller

But this is a strawman, as Casey Luskin pointed out on the Discovery Institute website: Response to Ken Miller & Judge Jones's Straw Tests

“Drs. Miller and Padian testified that Professor Behe’s concept of irreducible complexity depends on ignoring ways in which evolution is known to occur. .... Professor Behe excludes, by definition, the possibility that a precursor to the bacterial flagellum functioned not as a rotary motor, but in some other way... The ACLU speaking through Judge John E. Jones

Ah yes, Judge Jones bought the misrepresentation honest Ken gave under oath in Dover 2005, and honest Ken repeats it in the Nature review, but what Behe says shows that Miller is misrepresenting him:

Because the cilium is irreducibly complex, no direct gradual route leads to its production. So an evolutionary story for the cilium must envision a circuitous route, perhaps adapting parts that were originally used for other purposes.

This clearly shows Miller misrpresented Behe. Furthermore, it appears Miller hasn't been reading any of the ID press releases pointing this out. Golly, Miller's friends haven't been very good at sending e-mails to Miller to alert him of DI press releases with his name on it like Response to Ken Miller which point out his misrepresentations. So that's why honest Ken continues to repeat the misrepresentation in the prestigious scientific journal Nature (the same as he did in Dover under oath) after he has had the opportunity to correct his mistake for almost 2 years now. Even repeated it on camera in a debate with Paul Nelson, after Nelson told him in person at AEI in 2005, that Miller was misrepresenting Behe. Miller's so dense! Honest Ken is not dastardly, he would surely stop this misrepresentation once he was aware of his mistake. Rather, the reason he does what he does, even under oath in a major trial, is that he's comprehensionally challenged and dense. He's not dastardly in any way. No, not a chance. scordova

Ken Miller even mis-cites a paper:

(Nature Genet. 37, 73–76; 2005)

Actually it was:

(Nature Genet. 37, 73–76; 2004) The 'evolvability' of promiscuous protein functions

But this is more proof "honest" Ken is just sloppy and not dastardly. PS this paper that Miller References was also referenced in this peer-reviewed paper The intelligent design of evolution which stated

The debate between intelligent design and evolution in education may still rage in school boards and classrooms, but intelligent design is making headway in the laboratory. ..... Intelligent design, however, may be here to stay.

scordova

Sal, you should have been quoting me rather than Mike Gene, as I have priority on pointing out Miller's blatant falsehoods. His ignorance can only rightfully be viwed as willfull, and his mirepresentations as deliberate. Ever since I readi Finding Darwin's God I have doubted that he is a Christian, and not without reason. Mung

Miller states without proof whatsoever: First, he overlooks the existence of chloroquine-resistant strains of malaria lacking one of the mutations he claims to be essential (at position 220).

Did Miller forget this study? P. falciparum Digestive Vacuole Transmembrane Protein PfCRT and Evidence for Their Role in Chloroquine Resistance

All Old and New World pfcrt alleles in CQR parasites consistently include mutations for K76T and A220S

or even in the paper Miller cites and accuses Behe of waving off:

mutations K76T and Ala220 Ser (A220S) appear to be the most reliable markers predicting CQ resistance.

and the final nail in Miller's argument:

There are three plausible explanations for this: (i) If the mutations can be acquired in any sequence and K76T and A220S have large effects, then they will have a stronger correlation with resistance; the problem with this argument is that they [76 and 220] rarely, if ever, occur alone

PS For the situation where 220 is missing but we still find CQR, see comment on A144T/L160Y Behe was astute to note "the problem with this [sequentially selective argument] is that they rarely, if ever, occur alone". In other words, honest Ken is not a dastardly guy afterall, just comprehensionally challenged. scordova

//off topic It seems that all the big names in the darwinist camp will be commenting on Behe's book next month. and here is the first (Richard Dawkins no less) http://select.nytimes.com/preview/2007/07/01/books/1154680128921.html?em&ex=1183089600&en=03cc08833369a457&ei=5087%0A So, until now we have Michael Ruse, Sean Carroll, Jerry Coyne, Ken Miller and Richard Dawkins. I never thought Dr. Behe was that influential ;) IDist

Behe waves away evidence suggesting that chloroquine resistance may be the result of sequential, not simultaneous, mutations (Science 298, 74–75; 2002), Honest Ken Miller

yet in Behe's book which Miller supposedly read, Behe cites that very 2002 paper which Miller insinuates Behe waves away! See page 45 where Behe mentions "A requiem for chloroquine", which he references on page 280 as (you guessed it):

2002 Science 298, 74–75

Exactly the paper Miller says Behe waves away. Yet it is right there in Behe's book!!!! The paper may say that

The sequential accumulation of these mutations plausibly explains the observed genetics and epidemiology of CQ resistance

It doesn't actually prove it! What's this? Proof by bald assertion? Does that count for Miller? Honest Ken is not only ignorant he also has serious reading comprehension problems to be writing book reviews in the prestigious scientific journal Nature. scordova

Thanks for the post Sal. I have no access to the article :( But someone told me that Miller's argument is that you can have more than one mutation, none is advantageous, but when they all occur the organism gain an advantage. Is this really what he argues? Doesn't sound darwinian to me at all. IDist