Series reviewing Douglas Axe's Undeniable

Not only that, but their invisible being acting through unknown mechanisms somehow produced genomes and physical organisms in such a way that they look exactly like they evolved. Genomes and physical characteristics both fall into phylogenies that correlate with one another, exactly as we would expect from evolution. We see conservation of sequence in exons and genetic drift in introns. We see fossils with the exact mixture of features that we would expect to see in exactly the time periods we would expect to see them if evolution is true (e.g. Tiktaalik). None of this makes sense in Intelligent Design if a designer is constantly tinkering with organisms. Humans tinker with organisms all of the time, and we produce organisms that easily and obviously violate the phylogenies that evolution would produce.

1 Like

In Axe’s JMB paper he references several studies which suggest protein rarity:

  1. Yockey, H. P. (1977). On the information content of cytochrome c. J. Theoret. Biol. 67, 345–376.
  2. Reidhaar-Olson, J. F. & Sauer, R. T. (1990). Function- ally acceptable substitutions in two a-helical regions of l repressor. Proteins: Struct. Funct. Genet. 7, 306–316.
  3. Axe, D. D. (2000). Extreme functional sensitivity to conservative amino acid changes on enzyme exteriors. J. Mol. Biol. 301, 585–596.
  4. Taylor, S. V., Walter, K. U., Kast, P. & Hilvert, D. (2001). Searching sequence space for protein catalysts. Proc. Natl Acad. Sci. USA, 98, 10596–10601.

He also contrasts “forward” approaches which start with random sequences and look for function and reverse approaches which start with actual proteins and then look at their local rarity. What are the best reverse studies which demonstrate lack of rarity? Also, what are the best examples of forward studies which generate sequences which demonstrate stable folds with something close to a genuine active site? And, do problems exist with the studies Axe cites?

1 Like

Turns out the claim a stable fold is required is unwarranted. Also, especially if catalyst intermediate-stabilaization is understood, it’s clear that binding sites are conceptually equivalent to active sites. There is an immense number of studies from the 1.5 decades after your last ref that demonstrate selective binding is easy to find with far less resources that available in evolutionary history.

Moreover we know new function is exceedingly easy to evolve in mammals from cancer evolution.

The best reverse approaches are from analysis that takes coevolution into account, and thereby can capture interactions between mutations. There are wide range of approaches in this class, but they show the inference to rare function from mutational analysis of a few mutational effects function is incorrect.

Also very different sequences with the same structure or function are additiona evidence that the inference to functional rarity Axe makes is flawed.

1 Like

That seems false. Given what we know of neutral theory, God could have been constantly tinkering and we see the data just as we do. Science is silent on God action, and more over we know the number of mutations he’d have to tinker with would be vastly outnumbered by neutral mutations.

I’ve always wondered what Axe thinks about the formation of high-affinity antibodies in the vertebrate immune system. Specific binding isn’t a “function”?

1 Like

He would accurately point out its not an enzyme. However he would also incorrectly say that this a systme designed to produce high affinity, and therefore is not relevant

I would respond, the antibody system is just like evolution then, a system created by God to create new proteins.

I also remind him that the theory of intermediate stabilization predicts that antibodies, because the can adaptive ly bind things, can be catalysis. This is exactly what we find in Abzymes.

https://www.sciencedirect.com/topics/medicine-and-dentistry/catalytic-antibody

This demonstrates by theory and example that selective binding is functionally equivalent to catalysis. The reason we do not select for enzymes in experiments is because it is exteremely difficult for humans to select for catalytic activity at remotely the same scale we can select for binding.

Yep. A lot of my students, when they learn about antibody formation, make this connection as well.

Going with the scenario that God acted directly on genomes . . .

I would argue that there are nearly infinite number of genetic changes that would produce a clear and obvious violation of the expected phylogenies. God could swap out entire genes or even gene families between very divergent species, which is exactly what humans do in many cases. At least to me, it begs the question as to why God would choose not to move genes all over the species tree, and why God would instead opt for an extreme minority of mutations that would not run contrary to what we would expect from evolution. It gives the appearance that God made the extra effort to make any tinkering invisible to humans.

2 Likes

@Joel_Duff’s article is here. It is quite good. Very thoughful.

1 Like

Yes, well done @Joel_Duff. Romans 1 certainly seems to teach that all people have an “intuition” of God. There is plenty of proof for Paul’s assertion in the fact that all cultures throughout human history have believed that there is more to this world than we can see, as Joel pointed out.

Axe goes too far in asserting that this “intuition” takes a specific form, and – surprise! – that specific form is an argument from design. This reminds me a bit of Schweitzer’s criticism of the search for the “historical Jesus,” in which scholars examining the evidence tend to find a Jesus that looks a lot like themselves.

2 Likes

I had a wonderful conversation with a friend who is a biology professor south of me specializing in protein evolution. He gave some thoughtful feedback on the discussion of Axe’s work, so I will attempt to accurately convey all of his points along with some of my own thoughts.

One has to be very careful comparing different classes of proteins and generic claims of function. Many have the impression that proteins which function without fully folding into stable 3D configurations are suboptimal precursors to a fully folded, more evolved version. However this perspective is mistaken. The proteins in nature which function without a stable fold have actually “been optimized for multiple competing functional demands.” Their precise “looseness” is essential to their function, so they likely also represent extremely rare sequences of AAs.

Similarly, one must distinguish between any “function” and the specific functions performed by specific proteins which are needed for specific major adaptations. As an analogy, one could ask how likely would a child be able to craft a weapon out of a pile of garbage. The answer depends on what one means by “weapon.” Anyone could craft a sling shot out of a rubber strip and a few sticks. And, it might be useful for hitting a little brother ten feet away. However, the child would never be able to craft a high-precision riffle which Jason Borne could use to hit a target a mile away.

One of the most impressive experiments which started with random sequences was able to generate a chain which bound to ATP and catalyzed hydrolysis. The rarity of sequences for that function is somewhere around 1 in a trillion. I will refer to the chain as miniATPase. Let’s contrast this chain to the enzyme aconitase which changes citrate into cis-aconitate. This protein consists of over 700 amino acids compared to the less than 150 for miniATPase. I will focus on this example since it is part of the citric acid cycle which is a rather common metabolic pathway, and its activity is demonstrated beautifully by a YouTube video, which I encourage everyone to watch.

The enzyme performs several tasks:

  1. It binds to citrate, so the substrate resides at a precise location in a precise orientation. An neighboring Iron-Sulfur cluster helps stabilize the substrate electrostatically.
  2. Histone 101 donates proton to remove one OH group from the citrate. The enzyme is now altered.
  3. Serine 642 acts as a base by accepting a hydrogen from another location in the substrate. The enzyme is further altered.
  4. The interactions between the enzyme and the altered substrate flip the substrate upside down.
  5. Histone 101 binds to a proton from a passing water molecule, causing OH to attach to a new location in the substrate.
  6. Serine 642 returns a hydrogen atom to a new location in the substrate. The enzyme returns to its original state.
  7. The enzyme releases the substrate, so it can act on another citrate.

The successful conversion of citrate requires the right interactions, both chemical and electrostatic, to take place at the right times. Several amino acids have to be perfectly positioned, and the enzyme needs to have just the right stability, so the positioning is maintained, and the substrate can flip at the right time. Its activity only commences after the chain is perfectly folded. And, every step in the conversion process is essential. Since the enzyme alters during the first few steps, it must return to normal after the last steps, or it could reengineer other molecules. Therefore, it could not have evolved gradually, since the fold had to be highly optimized for all of the steps to proceed properly. The miniATPase example is like the slingshot from the analogy, and aconitase is like a high-precision riffle with a laser scope. Properly challenging Axe’s basic thesis requires one to focus not on the production of simple functions but on the most complex enzymes and other features required for novel adaptations.

Moreover, producing just once enzyme would not typically advantage an organism since most of the products of individual steps in metabolic pathways are useless without other enzymes to further process them. In fact, some intermediates are even toxic to the cell, so they have to be carefully ushered along to other enzymes. One cannot dismiss such challenges simply by referring to such comparatively easy tasks as antibody binding or the generation of other simple functions.

I asked my friend why critics of Axe’s research who specialize in protein evolution have not simply reproduced his experiment with what might be deemed as an improved approach. He responded by commenting that researchers who spend their careers attempting to create new protein folds or carefully study the properties of complex enzymes in nature know from everyday experience that actual enzymes like aconitase (not like miniATPase) are extremely rare. They instead simply believe that some pathway must exist to it from some ancestral protein, but this belief is based purely on faith.

My friend is very interested in this type of research. Would you please provide references and describe the specific examples to which you are referring? What is the most complex enzyme which was studied?

Also, one has be careful in assessing arguments made from comparing proteins in existing species, since they can easily fall into circular reasoning. Specifically, one could assume that undirected processes are responsible for the appearance of de novo genes or the differences between different proteins or other features in different species. And, then one could argue that the fact that those difference came about by natural processes demonstrates their power to create such features. The evidence presented depended on an assumption (the power of natural processes) that the evidence was meant to prove.

Happy to take this conversation offline with your friend directly. I’ll look forward to an email from him.

Well darn. I was looking forward to reading your response to this

2 Likes

I think Joshua @Swamidass will eventually come back with a response. I suspect he wants to avoid the kinds of communication problems that can happen when complex ideas get discussed in a forum. We have commentators here who suspect every ID proponent of being a closet YEC–or even worse, tolerating YEC views!–and therefore demand position statements on a variety of issues. Some folks manage to remain productive in such an environment, but others shut down. Why risk it?

1 Like

@bjmiller appears to be communicating the argument of a scientist that prefers to remain anonymous. I entirely understand why that is the case for those making ID arguments. They risking losing their careers over this.

However, arguments by proxy are not an effective way forward. When I’ve explained my position in private to others in the past, and it turns into an argument by by proxy, it has not worked out well either (just look at ID movement’s response to @vjtorley for an example). It was thought that I was hiding my identity on purpose, rather than choosing not to argue a highly detailed and technical point in public.

In this case, it is easy to see the problems in the argument from the get go. For those unclear, I’d point you to @vjtorley’s article on this which I largely endorse on the science: Undeniable packs a powerful punch, but doesn't land a knockout. I know Axe thinks that philosophers have no place in the conversation, but Vincent review requires response. For that review, where he quoted James Tour and myself at length, he was kicked out of Uncommon Descent (so much for the free exchange of ideas), and Axe dismisses it with an extremely disrespectful ad hominem in his response to a scientist.

This is fairly remarkable, because I cannot count the times I’ve been lectured by ID proponents about the need for scientists to listen to philosophers. So Axe’s rejoinder against @vjtorley (he is just a philosopher!) has heightened comic value for me. If that was a valid objection, Steven Meyers and Bill Demski (and the vast majority of ID sympathetic academics) should have just packed up and left this conversation a long time ago. Of course it is not a valid rejoinder, especially when the philosopher actually is explaining the science correctly.

Regardless, this is no different. It is legitimate to keep one’s identity anonymous in public, so it would make more sense to talk in private. I’m happy to engage and explain why Axe’s argument is entirely unconvincing, scientist to scientist. I am happy to talk in private about why this scientists rejoinder does not make the case. However, there is no real value in debating with someone not know who they are.

This is not a double standard. It is well known now that @vjtorley consulted with me on his article. If Doug Axe wants to engage with me in private about his work, and my objections to it, I am more than happy to do so. A public debate on a forum, that he himself is not even engaging, however, has very little value. In the past, when I have reached out to him, he has explained that he would rather take his argument forward with convincible non-scientists, than engage with me. So the lack of engagement here is his decision, not mine.

1 Like

The number of falsehoods in the piece was striking. That’s another reason the scientist should want to remain anonymous.

3 Likes

This is true too, but it is important to distinguish between falsehoods and lies. Most likely, this is just what this biology professor thinks, without knowledge it is false.

What follows is not a full engagement with @bjmiller’s post, but some of the key errors that are worth pointing out for the benefit of those following along.


Not really. It is correct, and not mistaken. We know for a fact that functions can appear in “disordered” proteins, and that these functions can become more efficient as the functional fold is stabilized. This entirely undermines Axe’s argument. The fact that some proteins (most in mammal’s?) never need to be fully stabilized for their function severely undermine’s Axes entire argument. He is working of an antiquated view (i.e. decade old, so maybe not antiquity) protein function. He would have to turn back the clock of our understanding to be correct.

Incidentally, it’s about when he left science that disordered proteins were beginning to be understood. We’ve just grown in our understanding since his last paper was published.

If he is unaware of this research (and there are several different approaches) he is certainly not an expert in protein evolution. There are several approaches, and literally hundreds if not thousands of papers here. No bluff here, he should contact me and we’ll talk about it.

On this I would totally agree!

For example, this is the exact error that Doug Axe’s argument falls into. He is arguing in a circular way. His argument essentially assumes that extant proteins are the only way to achieve what we see, so therefore they must be rare because extant proteins are only a small space of sequence space. That ends up being the fundamental error of his approach. Quite ironically, Doug Axe’s argument is just the type of circular reasoning that needs to be avoided.

However, the reason I raised the issue of co-evolution is to show just one reason why Axe’s inference to inaccessibly rare function by studying mutational tolerance of extant proteins is invalid. He grossly misinterpreted his data, ignoring some of the fundamental principles of protein structure.

Just false. There are several mechanisms we know of that can build up complex functions from simple functions. This also deviates substantially from Axe’s argument in the first place. Ignorance of evolutionary science is not an argument against it. One can dispute it, but ignorance is not a rebuttal.

That is a nice just-so-story with great anthropomorphic flare. We could tell the same story about Abzymes too. Literally no meaningful difference at all. Describing how a enzyme works in anthropomorphic terms does not somehow remove the fact that we can find enzymes in a forward approach. And of course, we could tell beautiful anthropomorphic stories about them too. That is why this sort of rhetoric is not convincing.

This is another fairly interesting equivocation worthy of discussion.

Notice what is assumed here? The argument implies that new complex enzymes are required for most (or all?) major adaptations. That, it turns to be false, and to be clear, that claim is only implied, not clearly stated. One of the great surprises of biology, however, is that this is not generally true.

Let’s take one of the most stunning strings of major adaptation in history, the evolution of humans from common ancestors with apes. Our intuition might tell us this would require evolving a large number of new and complex proteins. Our intuitions would be wrong (sorry Axe). As far as I can tell, there is no evidence than any new enzymes were needed at all. It turns out:

  1. To a first approximation, humans and the great apes have the same proteins, just turned on and off (expressed and spliced) in different patterns.
  2. To a second approximation, some of these proteins have tweaks that subtly alter their function.
  3. To a third approximation, a small number (<100?) of peptides (not enzymes as far as we know!) might possibly have arisen with subtle functions very hard to pin down. It is entirely possible all these de novo proteins are spurious and none of them are important.

So as surprising as this is, we there is no evidence (I can find) that any new enzymes were required for humans to evolve over the last 6 million years. There are no complex enzymes that we have that a great ape does not. Take that in. It is a really stunning fact. New enzymes are not required for the most important evolutionary development to us, the rise of humans.

If the evolution of humans is not a major adaptation, nothing else really is. What we find is that for large organisms (like mammals), most major adaptation takes place by just rearranging things. To make an analogy, you can make a lot of things with lego pieces (see lego sculptures). You do not usually need a totally new types of pieces for major adaptive change. The Lego analogy is a weak analogy, yes. However, in this one sense, it is true. It just does not take a lot of new types of pieces to make new things. We just need to arrange them differently. That is one of the grand surprises of comparative genomics.

With this fact in mind, at best Doug Axes argument, at absolute best, it is an argument against the origin of life and against bacterial evolution. It may be an argument against God-free evolution in cases where enzymes are important for major adaptations in mammals (where?). However, it is not argument against the God-free evolution of humans from a common ancestor with the great Apes. Remember, he does not engage any of the evidence for the common descent of man. So even if he is correct (and he is not), this reduces to an origin of life argument.

Of course, I am not arguing for God-free evolution. I understand evolution as God’s providentially governed way of creating us. If we allow for God’s providential work, this would not even be a good argument against the origin of life.

9 Likes

Thanks for taking the time to formulate the post, Josh. That was a great series of explanations showing exactly why Axe’s arguments are flawed.

This exact rebuttal was running through my mind this week. When we consider the creative power of God, the “arguments from odds” that Axe and others like to use are just plain moot. They are only applicable to a evolution without God, and not very effective in that regard, either.

1 Like

https://drive.google.com/file/d/1fymQV2zu50apoRfgc24b6pTdJDOrYXFb/view
I feel this may be relevant. Seems to address one of the claims made in the post

1 Like

Here is the next article, from the Faraday Institute:

That is why we see so many examples of what would appear to be bad design to an engineer. Some aspects of biology are like Heath-Robinson contraptions, rather than machines made by an intelligent engineer for dealing with specific problems.

Honestly, his argument about “bad design” ill advised. I believe God created us all, and He did not create us with a bad design, but through the process of common descent. I think rhetoric like this article on “bad design” just raises senseless confusion about God’s role in evolution.

However, there is some important thoughts on anti-intellectualism:

There is some implicit anti-intellectualism throughout the book and it is intensely worrying (“All Humans Are Scientists,” 60-64), and this way of thinking then turns in on itself as he attempts to use complex science to undermine established (and many would say proven) scientific concepts.

And here:

Axe’s approach is disturbingly close to the view of ‘who needs experts?’ that seems to be prevail in popular politics. Yet, of course Axe sets himself up as an expert, and his opinion is supposed to resonate with our own common sense/science.

It seems Axe wants to have it both ways. He wants to attack all the experts who disagree with him by making a populist appeal against experts and for “common science.” However, he also wants to turn around and dismiss critics like @vjtorley and, even other biologists, because they are not the real experts, like he apparently is. That is an incoherent strategy, but it is what he is doing.


Still, having read all the reviews, @Joel_Duff’s article is the best, by far. I highly recommend reading it i you have not yet seen it.

2 Likes