The Failures of Mathematical Anti-Evolutionism

Could you give some examples of these protein-protein interactions that are present in humans but not chimps (or vice versa)?

No, that is not a valid inference at all. They likely disappear, if they do, as a result of genetic drift.

BTW, how long does it take them to disappear? Is the time so short that simultaneous mutations are needed?

Another non-sequitur of a response. At the time Behe wrote his book, what reason did he have for assuming that the evolution of chloroquine resistance required multiple mutations?

That only applies to this one trait in this one organism. How does Behe generalize this to all traits in all organisms?

They are deleterious in that example because it is a made up example and Behe imposed the condition that four of the mutations are highly deleterious on their own.

If there was a trait that required eight mutations, all of which were neutral on their own (and in every combination) until all eight are present (at which point it is positively selected), Behe’s calculations will not apply at all. Do you understand that?


How about here? “Exon-1-encoded Sia-recognizing domains of human and ape Siglec-9 share only approximately 93-95% amino acid identity.”

“As discussed in Chapter 3, none of the changes seem to be improvements in an absolute sense. They disappear once drug therapy is discontinued.” (The Edge of Evolution, p. 136)

I gather from this that they disappear relatively quickly, i.e. not through drift, and such that simultaneous mutations are needed again.

Because the rate of chloroquine resistance (1 in 10^20) is about the square of atovaquone resistance (1 in 10^12), which requires one mutation.

He then turns to protein-protein interactions, and discusses how a new protein-protein binding site would be expected in about 1 in 10^20 organisms. He then places the edge of evolution at two new protein-protein binding sites.

But the mutations are independent events, and must occur together, so the probability of the 8 mutations is just the probability of each mutation, multiplied together. So I think Behe’s estimate does apply, in that sense.

No, you specified protein-protein interactions. Is Sia a protein? This is also about modifying specificity, well within even Behe’s fictional edge.


Neither you nor Behe has any evidence to support that.


OK, let’s take that as an example. Can you now demonstrate that this required four simultaneous mutations?

Is this just a guess on your part? Has it actually been demonstrated that every single mutation on its own is subject to negative selection? The Summers et al paper shows this is not the case.

Sorry, I misspoke there. I meant to say "multiple simultaneous mutations. In any event, we now know this is not the case because of the Summers paper.

I know he says that. How does he support that claim?

No, they don’t. They can occur separately, one at a time. Why not? You still have not answered this question, and I have asked it several time. I know Behe wants you to believe this. That is not good enough reason. He needs to provide a sound argument based on sound evidence. You still have not presented that.


Why must the mutations occur together, as opposed to occurring in sequence?


I doubt, based on seeing many examples of your mathematical inability, that you can compute the probability of drawing a full house.

Feel free to prove me wrong.

Don’t forget that the poker-player can redraw some or all of their cards.


You can. But before the fact, you don’t know which event to compute the probability for. Computing the probability of drawing a full house is futile if you then draw a flush instead.

All this talk of before-the-fact vs after-the-fact with poker is a smokescreen to hide that whenever you talk about biological systems instead of card games, you’re computing the probability of something that has already happened - because if it hadn’t happened you wouldn’t know what probability to compute - hence it’s after-the-fact.

Feel free to counter this by computing the before-the-fact probability of evolving something that did not actually evolve. Until then, you’re just bait-and-switching.


Well, how about here, then? “Hominoid-Specific De Novo Protein-Coding Genes Originating from Long Non-Coding RNAs”

Well, the de novo proteins were on average about 150 amino-acids long. It would be surprising if there was an all-selectable pathway to proteins of this length.

“We analyzed the characteristics of these de novo protein-coding genes. Consistent with previous reports [4], we found that the gene products were smaller, with a median length of 150.5 amino-acids, compared with 416 amino-acids in the human genome, suggesting the difficulty in de novo origination of long ORFs…”

“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. [9]” (The Edge of Evolution, pp. 50-51)

“9.Kublin, J. G., Cortese, J. F., Njunju, E. M., Mukadam, R. A., Wirima, J. J., Kazembe, P. N., Djimde, A. A., Kouriba, B., Taylor, T. E., and Plowe, C. V.2003. Reemergence of chloroquine-sensitive Plasmodium falciparum malaria after cessation of chloroquine use in Malawi. J. Infect. Dis. 187:1870–75; Cooper, R. A., Hartwig, C. L., Ferdig, M. T. 2005. Pfcrt is more than the Plasmodium falciparum chloroquine resistance gene: a functional and evolutionary perspective. Acta. Trop. 94:170–80. Drug resistance mutation in pfmdr, the other protein involved in chloroquine resistance, also incurs a fitness cost (Hayward, R., Saliba, K. J.,Kirk, K. 2005. pfmdr1 mutations associated with chloroquine resistance incur a fitness cost in Plasmodium falciparum. Mol. Microbiol. 55:1285–95).”

“In 1993, Malawi became the first African country to replace chloroquine with sulfadoxine-pyrimethamine nationwide in response to high rates of chloroquine-resistant falciparum malaria. To determine whether withdrawal of chloroquine can lead to the reemergence of chloroquine sensitivity, the prevalence of the pfcrt 76T molecular marker for chloroquine-resistant Plasmodium falciparum malaria was retrospectively measured in Blantyre, Malawi. The prevalence of the chloroquine-resistant pfcrt genotype decreased from 85% in 1992 to 13% in 2000. In 2001, chloroquine cleared 100% of 63 asymptomatic P. falciparum infections, no isolates were resistant to chloroquine in vitro, and no infections with the chloroquine-resistant pfcrt genotype were detected.” (from here)

I think this implies both mutations are subject to negative selection.

The rate of chloroquine resistance is about the square of the rate of atovaquone resistance, which requires one mutation. This is what we would expect, if chloroquine resistance requires two independent, singly-non-selectable mutations.

How so? As far as I have read, they say there are two known paths, with two mutations required in each path, for basic resistance.

I’ve posted his argument in these threads, it has to do with considerations of protein shape space. I refer you to Behe’s book, if you want the full argument.

But they won’t last long, if they are singly somewhat deleterious.

My point is that viewed before the fact, the probability of an event can be less than 1, whereas after the fact, it is 1 or 0. So we have to pick one perspective, and stick with it. I choose “before the fact”, because that’s really the probability of interest, in card games, and in biological systems.

IOW “These go up to 11.” And just what does the length of the protein have to do with this?

I will repeat the question one more time: Why can the mutations necessary for these new(ish) proteins not arise one at a time? They don’t have to each be selectable. But even if they did, they would not have to occur simultaneously. Why would they?

Why do you think it implies that? One thing you do not seem to appreciate is that chloroquine would exert a strong selective pressure in favour of resistant strains. Once the drug is withdrawn, this pressure no longer exists and other variants are free to arise. It is not necessary that the resistant strain be subject to negative selection. But that is not the main problem you have here, so I just add that for your edification.

You really need to explain why you think these traits require simultaneous mutations.

Yes. If you insist it requires simultaneous mutations, please quote where this is said in the paper. Please, no more Nigel Tufnel. Give us the quote.

No, you really haven’t. You just think you have. “These go to eleven” is not an argument, sorry.

Again, support this assertion. And how deleterious is “somewhat”? Have you heard of nearly neutral theory?


No, your claim is about protein-protein binding sites. You’re very bad at inferring content from titles and abstracts.

You’re terminally confused.

The intermittent selection, even if all patients took chloroquine, is the reason why fixation of mutant HAPLOTYPES is so rare, not mutational requirements.

And that’s overall. It just as easily could be that some of the alleles that confer chloroquine resistance have decreased fitness in the intermediate hosts irrespective of chloroquine.

[quote=“lee_merrill, post:80, topic:15042”]

The rate of chloroquine resistance is not merely a function of mutation, but of intermittent selection, recombination, epistasis, etc. Behe’s “calculation” ignores all of these important factors.

You did not read carefully. There are two MAIN paths. You are misrepresenting the bit of the paper you quoted.

[quote=“lee_merrill, post:80, topic:15042”]

No, it does not, as there is nothing that sophisticated in Behe’s book. You’ve been bamboozled. Again, high-affinity, specific binding is evolved from random variation in your immune system in only 2 weeks.


I’ve restored part of my post that was omitted by @lee_merrill when he responded:

So you’re not going to calculate any before-the-fact probabilities of anything other than what you know evolved. All your biological system calculations are done after-the-fact, and your claims otherwise are untrue. You are bait-and-switching.


If Behe is basing his probabilities on what evolution does then he is picking after the fact.


Yes, that is another problem with Behe’s argument that we have not discussed at length here. This chart is from the paper entirely on which Behe based his 10-20 number1, and lists the conditions that must be met before a chloroquine resistant strain is detected in human beings:


Behe ignores points 2-8 and pretends the frequency is accounted for entirely by factor 1. That, by itself, ought to be sufficient to sink his argument if he had one. But the other point that is easily missed is that Behe does not even have an argument. He just has a series of unrelated points, each of which is true to varying degrees, and then he magically conjures a conclusion from them. Try map out his argument and see if you disagree. In fact, I might do that myself…

1JCI - Antimalarial drug resistance


The longer the protein, the more unlikely it is to have an all-selectable pathway to it.

They don’t have to occur simultaneously, but if one or more of the mutations required, is non-selectable (which would seem likely, since most mutations are actually deleterious), then multiple mutations would need to occur basically simultaneously.

It’s a deduction, first a minimum of two mutations is required: “A minimum of two mutations sufficed for (low) CQ transport activity, and as few as four conferred full activity.” Then, based on the rate of occurrence of chloroquine resistance (1 in 10^20) versus atovaquone resistance (1 in 10^12) we can deduce that chloroquine resistance requires two simultaneous mutations.

Deleterious mutations are selected against, why do I need to support this assertion? And chloroquine resistance mutations disappear in a few years, which would seem to imply that these mutations are not just neutral.


Human proteins are littered with neutral mutations that have accumulated over hundreds of millions of years. The two papers @Rumraket has linked to show how there are multiple pathways that utilize neutral mutations, so they are plentiful. Also, there are many human proteins that have known mutations to almost all non-lethal positions in the protein within the current population.


You all are misunderstanding me, I’m saying that when you are evaluating a probability, you have to pick before-the-fact or after-the fact, regardless of whether the event has actually occurred. All the interesting probabilities are deduced by viewing the event before-the-fact, so everyone will generally focus there.

I was assuming that it is not uncommon for proteins to bind to other proteins.

I’m not sure why you emphasize haplotypes, here? And Behe’s point is that the mutant appears to be weaker, that’s a pretty plain point.

Again, Behe is not calculating, he is observing what evolution does.

“Multiple mutational pathways led to saturable CQ transport via PfCRT, but these could be separated into two main lineages.” (Summers et al.) It sounds like “separated” implies total division into two lineages.

So how do you calculate the number of possible combinations of two mutations, one neutral and one beneficial, that can result in a selectable function in a given genome?


Proteins do not need an all selectable pathway, though. Why do you keep saying they do?

No, not if they are neutral. Which, contrary to your claim there, most of them are. Your problem is you just swallow everything Behe tells you, and he is usually wrong about what he says.

Major logic fail. Even if the Summers paper didn’t actually tell you why CR arises infrequently as it does, your reasoning would still be invalid. There are any number of reasons that a particular trait might arise infrequently besides a requirement for simultaneous mutations. But as it happens, the paper actually tells you why the resistant strain is less frequent than might be expected. Note the bolded sentence:

Mutations in the chloroquine resistance transporter (PfCRT) are the primary determinant of chloroquine (CQ) resistance in the malaria parasite Plasmodium falciparum . A number of distinct PfCRT haplotypes, containing between 4 and 10 mutations, have given rise to CQ resistance in different parts of the world. Here we present a detailed molecular analysis of the number of mutations (and the order of addition) required to confer CQ transport activity upon the PfCRT as well as a kinetic characterization of diverse forms of PfCRT. We measured the ability of more than 100 variants of PfCRT to transport CQ when expressed at the surface of Xenopus laevis oocytes. Multiple mutational pathways led to saturable CQ transport via PfCRT, but these could be separated into two main lineages. Moreover, the attainment of full activity followed a rigid process in which mutations had to be added in a specific order to avoid reductions in CQ transport activity. A minimum of two mutations sufficed for (low) CQ transport activity, and as few as four conferred full activity. The finding that diverse PfCRT variants are all limited in their capacity to transport CQ suggests that resistance could be overcome by reoptimizing the CQ dosage.

Got it? The mutations had to be added in a specific order, but were still added one at a time. The astonishing thing is that it says right there that Behe is wrong, yet he and his gullible sycophants actually thought this paper vindicated him! I find that very strange, and very sad. Don’t you?

Jeez, you really don’t read very carefully. Go look at what I actually asked, then see if you can come up with an answer.

"In a few YEARS??!!! " And you honestly think drift alone is not capable of removing a variant that is not subject to selection either way in an organism that reproduces as rapidly as C. falciparum in a matter of a “few YEARS?”

This is the problem: You really have no understanding of the bare basics of this subject, so you are vulnerable to Behe’s underhanded efforts to bamboozle you with his BS because he is saying something you want to be true. I actually feel sorry for you, because then you come onto a forum like this and embarrass yourself.