[00:00:00] Speaker A: Foreign.
[00:00:05] Speaker B: The Future, a podcast about evolution and intelligent design.
[00:00:11] Speaker A: Welcome to ID the Future. I'm your host, Eric Anderson, and I'm pleased to have Rob Stadler back on the show with us today to continue our discussion about confidence in science. Welcome, Rob.
[00:00:21] Speaker B: Hi, Eric. It's good to be here. Thank you.
[00:00:23] Speaker A: You bet. Appreciate having you back on.
So last time we had talked about a book that you wrote and some criteria that you had identified to allow us to determine whether we're dealing with high confidence science or low confidence science without going into them. Just, just list them, encourage viewers to go back and listen to the prior episode, watch the prior episode, but just list for us real quickly what those six criteria were, Rob.
[00:00:49] Speaker B: Yeah, and also a reminder that these are not meant to be just simply yes or no. It's a spectrum of confidence. You could be somewhere in the middle or middle, something like that.
[00:00:58] Speaker A: Yep, good point.
[00:00:59] Speaker B: But we want to know if the, if the evidence is something that is repeatable, can you repeat the findings? Can you directly measure the result or directly observe it, or is it very abstract?
And was the evidence obtained through a prospective study that you had planned in advance and then you conduct the study so you can control any kind of confounding factor?
[00:01:20] Speaker A: Yeah.
[00:01:21] Speaker B: And was, was bias minimized somehow?
Actively.
And were assumptions minimized? And if you take any assumptions, do you openly disclose those and justify them?
And then finally, do you make reasonable claims about the results that you obtained under the conditions that you obtained them?
[00:01:44] Speaker A: Okay. And on that last point, that sixth one about making reasonable claims, you had shared a quote from a biology textbook that's very widely used.
Why don't you share that with us again and then go ahead and speak to that for a bit?
[00:01:57] Speaker B: Yeah. So we're now diving into applying these criteria to evolution. And when I read my son's high school biology textbook, you know, this, this one particular claim jumped out at me and in the book, and it says, for example, organisms as dissimilar as humans and bacteria share genesis, inherited from a very distant common ancestor.
And we talked last time about how that, that claim seems to be exuding confidence. There's nothing there like, you know, we suspect or evidence suggests or, or, or
[00:02:32] Speaker A: some scientists believe or, you know, anything like that.
[00:02:35] Speaker B: It's a very firm statement. And, and I thought we would talk through that and, and look at the levels of confidence that, how, how it relates to these six criteria for confidence in that one statement.
[00:02:48] Speaker A: Yeah, good, good.
[00:02:49] Speaker B: But I also wanted to point out more generally that with, with that statement, we're referring to evidence for evolution called homology is. Is the fancy term for it, but it's basically just saying that we see a lot of similarity between different kinds of animals. It could be like the morphology of your limb, that you have one long bone and two long. Yeah, two bones for forehand.
Or it could be the similarity of. Of your DNA, of the code or the genes, as this particular quote from the biology textbook is focusing on, that you share genes.
So that's all. All under the umbrella of homology.
And I wanted to play this, this snippet from a YouTube video from Richard Dawkins where he's directly asked what is the most convincing evidence for evolution? And.
[00:03:40] Speaker A: Okay, yeah, let's play it.
[00:03:41] Speaker C: He says, out of all the evidence used to support the theory of evolution, what would you say is the strongest, most irrefutable single piece of evidence in support of the theory?
There's an enormous amount of evidence from all sorts of places, and it's hard to pick one strand which is more important than any other.
There's fossils, there's the evidence from geographical distribution, there's the evidence from vestigial organs. I think, to me, perhaps the most compelling evidence is comparative evidence from modern animals, particularly biochemical comparative evidence, genetic, molecular evidence.
[00:04:18] Speaker B: Okay, so there you. You heard Richard Dawkins say that the most irrefutable and the strongest evidence that they have for evolution is. Is basically homology. It's basically comparing between life forms and looking at what I would call the correlation structure, you know, between one life form and another. You see a high correlation between gorillas and chimps, between chimps and humans. You can see this correlation structure, and they speak to that as very strong evidence for common ancestry. So the simple question is then, how does that hold up against these six criteria that tell us if we can be confident in a claim?
So we have the six criteria. First of all, I always find it best to when. When you have a scientific claim, you kind of have to play a little bit of Jeopardy where you turn the claim into a question.
[00:05:11] Speaker A: Okay. Yep.
[00:05:12] Speaker B: So I turn this into the question of do humans and bacteria share genes inherited from a very distant common ancestor? That's the question you're asking of science.
And first you say, well, is that evidence repeatable?
And whenever I bring that up, people say, yeah, yeah, it's repeatable because I can look at the genes in bacteria and I can look at the genes in a human, and I can see that they're the same. Right. There's commonality to some shared genes. There well, when they say that, they're actually answering a different question.
Now they're answering the question, do humans and bacteria share genes? Right, okay, yeah, of course we share genes, and you can say that very confidently. But the question we're trying to answer is do you share genes inherited from a very distant common ancestor?
[00:06:00] Speaker A: So you're talking about. You're talking about a causal event rather than the mere observation of similarity.
[00:06:07] Speaker B: Right. So why is it we share genes?
[00:06:09] Speaker A: Yeah.
[00:06:10] Speaker B: What's the cause of us sharing genes?
And that is something that you can't repeat.
You could. You can imagine if you had the common ancestor and you brought them into a laboratory and then you watched generation after generation of inheritance handing down genes, and that evolves into bacteria, that evolves into a human. You know, then you've got some pretty high confidence in that claim. But as you could tell, it's very, very far from something you could repeat.
[00:06:37] Speaker A: All right, well, that's pretty high standard, Rob. You're asking a lot of us here as proponents of evolution. So what about some of the other criteria? Maybe we can.
[00:06:46] Speaker B: Yeah. Number two, can the evidence be directly measured or observed? Again, you can directly measure the DNA in a human and the DNA in bacteria and see that there's similarities there. But that's not answering the right question. Right. The right question is, did they get those from a common ancestor?
And that most certainly can't be directly measured or observed. It's something way, way back in time.
And number three, can you study this prospectively? That would be the kind of scenario that I mentioned of going into the laboratory and watching them evolve.
[00:07:18] Speaker A: Can't possibly do that as a practical matter. No.
[00:07:21] Speaker B: And is bias minimized? I mean, I think statements like this are almost from bias, that the voice of the bias overpowers the voice of any relevant evidence.
And certainly people are on both sides of this with bias.
And were assumptions minimized or openly disclosed?
I find this statement, actually, it basically is an assumption because we have genes and bacteria have genes that are in fact, similar. We assume where they came from.
It is an assumption. And then, number six, did they make reasonable claims? Well, that we already talked about.
There's no hedging language here. It's a very strong, confident statement.
So I believe, I think they got all six of these wrong.
And so this counts as a very low confidence evidence kind of a statement, Right?
[00:08:16] Speaker A: Yeah, I'm thinking about this kind of from just a logical standpoint. This sentence here from this biology textbook, it's in some ways almost just a restatement of the theory, rather than an evidence based observation that has really gone through a rigorous analysis. I mean, the assumption is that we, you know, came from a common ancestor and share similarities that we attained from a common ancestor. That's sort of the theory. And so we're just kind of restating it here and using the word gene for this particular criteria that we're interested in. But it's kind of just a restatement of the theory.
[00:08:50] Speaker B: Exactly. It seems quite circular.
[00:08:53] Speaker A: Yeah, yeah. Okay, but all right, you know, and
[00:08:56] Speaker B: this is specific to this one statement, but sure. You know, what Richard Dawkins said in, in that video, that this is really the best evidence they have for evidence and for evolution. And I find that kind of striking, you know, in the. You take the best evidence you got and it happens to fail all six of these criteria, then you have to wonder, well, what, what's really going on here? And, and is there any evidence that actually provides high confidence that should be prioritized?
[00:09:24] Speaker A: Right, right. And, and just for our viewers, again, homology, we often, like you say, think of something like the pentodactyl limb, but it also can be genetic, which is what this particular claim was. Yeah.
[00:09:34] Speaker B: Yep.
[00:09:35] Speaker A: Okay.
All right, so what now? What do we make of this?
[00:09:42] Speaker B: Well, I just want to see people apply these criteria and think through them and to appropriately prioritize evidence that gives high confidence over evidence that can only give you very low confidence. And it's not meant to be, you know, kind of black and white. Like I said, it's a spectrum of different levels of confidence.
But as we saw in this situation using homology to try to prove macro evolution, it is actually a very, very low confidence level of evidence.
[00:10:14] Speaker A: Right, okay, well, that's not the only evidence for evolution. Right. We could look at the fossil record or something like that and say, hey, yeah, maybe this homology with the genetics isn't quite as strong in terms of evidence, but we got the fossil record. We're dealing with real bones here. We're dealing with real physical evidence that, that we can look at and see and hold and touch.
[00:10:34] Speaker B: Yeah, well, it turns out if you, if you run that fossil evidence through these same six criteria, you pretty much get the same result that, that we just went through. Because when you take fossil evidence, once again, like, you have to play Jeopardy and turn it into a question. Okay, so you're going to try to answer a question like, like did dinosaurs evolve into birds?
You know, or some kind of question like that. You ask of evidence of evolution, you ask that of science, and you say, is the evidence repeatable?
Well, you can't take a dinosaur and repeatedly evolve it into a bird. And you can't directly measure a dinosaur evolving into a bird, and you can't prospectively study that happen.
And of course, there's a lot of bias in this whole topic on both sides. And are there assumptions? You know, lots of assumptions there. Some people love to use the assumption that in a given fossil layer, if you don't find an organism, that means that organism didn't exist in that fossil in that time of. In that history. Right.
But often also you'll see people say, well, where, where are the transitional fossils? Where are those sure thing. Transitional fossils? And they'll say, well, just because we don't find the transitional fossil doesn't mean it didn't exist.
It just means you didn't find. So those are both assumptions and they happen to be contradicting each other.
There's a lot of assumptions here.
And then did they make reasonable claims? Is number six.
And when it comes to fossil evidence, here's the challenge is you're trying to assert how a living thing came into being.
[00:12:19] Speaker A: Right.
[00:12:20] Speaker B: What is the reason why this thing is alive?
And the evidence that you have is actually bones, you know, buried. So you have evidence that the thing did live.
[00:12:32] Speaker A: Right.
[00:12:32] Speaker B: And you have evidence that it died. Those things I can say for sure.
But to say that A evolved into B.
Yeah.
You're trying to explain how life came about, but the evidence you have is bones.
[00:12:46] Speaker A: Yeah.
[00:12:47] Speaker B: So the evidence is over here, far on the right. And what you're trying to explain is something far on the left.
And that is not, I would say, a reasonable claim. You're extrapolating, you're extending it beyond the data you actually have.
[00:13:00] Speaker A: Yeah. That's interesting. I actually ran into this a few days ago when Brian and I were out at BYU after one of his lectures. He actually gave three lectures there. But after his afternoon lecture, there was a gentleman, very thoughtful scientist there at the university who said, well, look, we've got this fossil record that shows this transition from land mammals to whales. And so, you know, what do we make of that? How are we supposed to deal with that?
It seems like it's pretty good evidence. And so, first of all, I pointed out to him that if you actually dug into the alleged transitions, it's not quite as impressive as it's usually made out to be. And I know you've done a lot of work and looked at that as well. We have some great videos from discovery on that.
But secondly, the point that I was really trying to emphasize to him is even if you assume that you have this beautiful series of fossils, the question is, what is the causal event here? What caused this A to turn into B, like you say, to C on up to Z?
And that's an aspect that the bones tell you really nothing about.
[00:14:05] Speaker B: Right.
Very well said. Yes.
Like I've said before, it's easy to come up with associations.
[00:14:13] Speaker A: Yeah.
[00:14:13] Speaker B: You know, this fossil seems to have feathers. This fossil seems to have feathers. So they're associated in that regard. But to say that A evolved into B is a causal state.
[00:14:25] Speaker A: Yep.
[00:14:25] Speaker B: And that's a big stretch from what data you have.
[00:14:29] Speaker A: Right, right. Especially when any sufficiently large database will give you associations and hierarchies. I don't care what, I don't care if you're talking about, you know, planetary bodies or lakes or mountains or technology, any sufficiently large set of objects will give you a hierarchy and associations. So it's pretty easy to find similarities and make up a story about them.
[00:14:54] Speaker B: Well said. And it's very difficult, in contrast, very difficult to prove out causality.
[00:14:59] Speaker A: Yeah. Yeah. Okay. All right. So there's been.
So we're not too impressed then with the level of confidence. And again, it doesn't necessarily say this cannot be true, but we're not impressed with the level of confidence that we have on something like the claim about the common genes or the homology of the genes. The fossil record isn't giving us a lot of confidence as well. Is there some high confidence evidence related to evolution that people should know about?
[00:15:28] Speaker B: Yeah, there is.
[00:15:29] Speaker A: Okay, tell us a little bit about that or dive into it. Let's go ahead and jump in.
[00:15:34] Speaker B: Let's dive into it. Okay, so I have here on this slide, there are three studies I'd like to talk about that do in fact meet all six of these criteria. And they give us high confidence of the process, of what the process of evolution can accomplish. Okay, and.
And it bothers me that textbooks got this upside down. They speak a lot about the low confidence evidence, like the fossil record, entomology, and they don't speak about the high confidence evidence because it's telling a different story if it's interpreted appropriately.
So three studies here.
The first study is from Ann Gager, and this was published in 20, 2010. And it's about bacteria's ability to produce its own tryptophan, which is an amino acid that it needs in order to survive.
And it uses these enzymes to produce tryptophan. And so they, they studied it and they said, hey, let's, let's give evolution a test.
We're going to cause a purposeful mutation. We're going to do this on purpose. Genetic engineering. We're going to create this mutation prospectively. Prospectively, that's fine. And repeatedly, and see if evolution can fix the problem that we introduce.
[00:16:54] Speaker A: Okay.
[00:16:55] Speaker B: And thinking, you know, certainly if evolution produced this entire gene, you know, from scratch or even borrowing parts from somewhere else, whatever it was, it produced this gene and this enzyme, if it can do that much, it should be able to fix some very minor errors that we introduce into it. Okay, so in the first part of the experiment, they changed one letter of DNA, just one nucleotide, which damaged the function of this enzyme, trying to produce tryptophan. And they put the bacteria into a media, a bath, that has a limited amount of tryptophan. So they create a natural selection, you know, selecting agent that if you can't produce tryptophan, you're going to lose, you know, you can produce it, you're going to win.
So they set this all up and they let the bacteria evolve over, you know, a couple of generations is all it took. And after about a hundred million E. Coli were grown in this media, it would seem that a random mistake in the DNA in replication fixed that single wrong nucleotide.
[00:18:04] Speaker A: Okay.
[00:18:05] Speaker B: So it was able to produce tryptophan again. And that one took off, took over the population and won, won the game.
So that tells us that evolution has the ability then to, you know, with enough numbers of organisms, it could fix by random chance, fix a kind of a single point mutation.
[00:18:23] Speaker A: And the numbers, if you run through those, were consistent with the typical mutation rate and the size of the population. So that you could feel at least somewhat confident that this was, or at least it could be explained by, let me say it that way, which is even more high confidence. One reasonable explanation would be that this was a random mutation given the population size and the general mutation rate. Correct.
[00:18:47] Speaker B: Yeah. They often throw out the number of like one in a hundred million nucleotides that's copied will be copied wrong.
[00:18:54] Speaker A: Right.
[00:18:54] Speaker B: And so this is about right. Yeah, makes sense. Okay, so they did a second time, they did here, they changed another single letter of DNA, but it was a different one, and then this time the enzyme now had no function. Instead of just being damaged, it was just not functional.
[00:19:13] Speaker A: Okay, okay, so sorry, just to clarify, the prior one could produce some tryptophan, but not to the level of efficiency that it normally does.
[00:19:21] Speaker B: That's right.
[00:19:21] Speaker A: Okay.
[00:19:22] Speaker B: Exactly. And this one was more damaging, so it's now not able to produce at all. And then they did the same things. They grew that mutant E. Coli with one bad.
One bad letter of DNA. They grew it in a population in a solution with a limited amount of tryptophan, and they basically attained the same result that after about 100 million organisms were grown in that media, one of them was lucky enough to solve the problem, fix that little bug, and then start to produce tryptophan, and then took over the population. Okay, very cool.
And then they did a third experiment, and in the third experiment, they introduced both of those errors. So now we have two bad nucleotides.
[00:20:05] Speaker A: Yep.
[00:20:06] Speaker B: And again, the tryptophan enzyme wasn't able to produce tryptophan, and they put it in the solution, again with a limited amount of tryptophan. This. This mutant who had two errors, but this time they got a really different answer because they went through, like, 9,300 generations of E. Coli grown in this media, and now we're talking about a massive amount of organisms total. It's like a trillion E. Coli organisms, which, when you think about it, that's. That's more than human beings ever have produced.
[00:20:41] Speaker A: Right.
[00:20:41] Speaker B: In humans and. And probably ever will produce, you know, because we don't produce that much.
And what they found after going through all that was not only did the E. Coli not get back to producing tryptophan, but the genes for doing it were either deleted, you know, just discharged, or they were just suppressed. And they believe that's because, you know, producing. Trying to produce enzymes that aren't functional actually uses tryptophan. And a tryptophan is a limited, precious resource, and you're wasting it by, you know, trying to produce enzymes that aren't helping.
So you're better off to just get rid of it, you know, give up.
[00:21:21] Speaker A: So this is pretty remarkable. Let me just make sure I'm understanding it correctly. So you're saying that with.
Was it the same two mutations that they had previously worked on, or was this different ones, the same ones, and none of the population were able to re. Achieve normal tryptophan production?
[00:21:40] Speaker B: That's right.
[00:21:41] Speaker A: That's pretty remarkable because I would have thought that potentially even, you know, in your second example, you had the gene completely broken, basically, it wasn't able to produce tryptophan correctly. I would have thought that if you had somehow been able to fix that, then you'd at least be back to the A scenario, and then you could potentially Fix that.
[00:22:03] Speaker B: So, yes, you're right. The key point is if they fixed one of the mutations, there's a little bit of a survival benefit to getting that one fixed. And they thought in the study that that would help it along to get the. Sure one fixed.
[00:22:15] Speaker A: Yeah, well, that's. Yeah, one would. One would. I mean, that's certainly a reasonable thing to think, but what you're saying is that the.
Even with this massive population size and the number of generations having these two mutations was. Was a barrier.
[00:22:33] Speaker B: Yeah. It's basically devastating. You know, and, and to put it into some better numbers here, this tryptophan synthase gene actually is actually two genes to make this enzyme.
And if you look at the number of nucleotides in those two genes, it's almost 2000. 2000 nucleotides. And they actually just had two mistakes out of 2000. So that's 99.9% accurate.
[00:23:01] Speaker A: Yeah.
[00:23:02] Speaker B: Genes.
And yet evolution couldn't push it to the finish line. You know, you're at, you're at 99.99 point now.
[00:23:09] Speaker A: You, you're. You're at the, you know, the, the one yard line. I just got to push it over the. Over the line here.
[00:23:14] Speaker B: Yeah. And it. And it basically tripped and fell and couldn't cross the finish line from that starting point. And. Okay, but yet, but yet we're told, this is what I'm taught in evolution class in school, was that all of this developed through the process of evolution.
[00:23:30] Speaker A: Yeah. You can go the whole hundred yards without any problem. It's not. Not a big deal. But it turns out you can't even get to the finish line when you're right at the goal line.
[00:23:38] Speaker B: Right. And actually, Eric, it's worse than that because tryptophan is actually produced through five different enzymes, and it's actually more than that. But you start with this reagent chorismate, and you go from there with five different enzymes to get to tryptophan.
And we just talked about one of those enzymes. We look at all five as a package deal, apparently. Assembly lines.
[00:24:02] Speaker A: Yeah, it's.
[00:24:03] Speaker B: And you got about 6,800 nucleotides in this whole assembly line.
And now you're saying two, only two were bad out of 6,800.
[00:24:13] Speaker A: Yeah.
[00:24:14] Speaker B: So it's 99.97% accurate.
And the process of evolution, supposedly random mutations and natural selections could not carry it over the finish line. It basically said, I'm out of here. I quit.
[00:24:27] Speaker A: Yeah. And it sounds like. And I know you weren't Necessarily talking about Mike Behe's work here, but it sounds like the result was consistent with what he found in some of the things that he studied. Meaning if I'm not getting this immediate survival advantage, I'm more likely to just break or blunt or delete this gene that's costing me resources rather than trying to get over the finish line.
[00:24:49] Speaker B: Yep. Just probabilistically taking a shortcut for, for a short term benefit is better than working hard to have the long term benefit.
[00:24:59] Speaker A: Okay, all right, so let's, let's. So this is fascinating research.
It was prospective in that they designed it and had the conditions up front. Right.
Talk us through the other criteria here.
[00:25:11] Speaker B: Yeah, it actually, it's repeatable.
[00:25:12] Speaker A: Repeatable.
[00:25:13] Speaker B: 12. 12 different replicate population.
[00:25:15] Speaker A: Oh, they did. Okay.
[00:25:16] Speaker B: That they worked on. And of course you could do this experiment yourself and repeat it.
And it's prospectively measured. They directly measured. Prospectively designed. They directly measured the results. Of course, you look at the genome of the E. Coli before and after, you could see the changes that happen.
Bias, you know, you could accuse them of bias. I suppose so. I wouldn't say that, you know, they didn't make a million dollars off of this.
Yeah, that's always a possible accusation, Assumptions. I don't know what assumptions.
[00:25:50] Speaker A: Well, the only assumption I can kind of see in here is, well, it depends on what their claims were and how broadly they made their claims. If they kept their claims appropriately narrow, then not really. But if you're making a broader claim about evolution, the one assumption in here would be that this is the kind of thing that we ought to see taking place generally in evolution. And if this doesn't work, then maybe there's a problem broader for the theory. Right. I mean, that's, that's sort of an assumption. It seems like a pretty reasonable assumption given that everything according to evolutionary theory came about in some kind of a similar manner. But that would be one. That'd be the only place where I could say, okay, great, can we redo this experiment with maybe a few different enzymes and maybe that'd be some good further work to do.
[00:26:36] Speaker B: Right. And even, even in a different kingdom of life.
[00:26:39] Speaker A: Right.
[00:26:40] Speaker B: To repeat the experiment, you know, but
[00:26:42] Speaker A: this seems like a good place to start. Okay.
[00:26:44] Speaker B: Yep.
[00:26:45] Speaker A: And then what was their, what was their claim? Do you. I mean, they're just reporting the results or are they making any particular.
[00:26:51] Speaker B: Yeah, I don't, I don't have the paper right in there to pull up. But. But I don't think they, they claimed anything beyond that. They didn't say, you know, this proves that all evolution is incompetent to solve any problem or something. They just. They just kind of laid out the results. So.
[00:27:05] Speaker A: Okay. All right. Well, that's pretty interesting research.
[00:27:08] Speaker B: Yeah. And I think, you know, compared to that claim that we just discussed about homology and about shared genes between humans and E. Coli with our common ancestry, you know, this one clearly is a higher confidence prospective study, and I think it needs to be prioritized.
[00:27:25] Speaker A: Good, good.
[00:27:26] Speaker B: Now we can move on there. There's three of these I wanted to go through. So the second one is actually in yeast.
So it's kind of like I just had mentioned, this is a different kingdom of life. We are now into eukaryotes, we have yeast, and we have a pretty similar result, actually.
The paper shown here came out in 2021.
It's called phenotypic and molecular evolution across 10,000 generations in laboratory Budding Yeast Populations.
And I'll say that the part I'm going to cover is not actually the main point of the study. It's kind of a sidetrack that happened as they went through the study that I find really fascinating.
So the yeast that they studied, it's a strain of yeast called W303, but it has a known mutation, a single point mutation in it in the process, in the biological pathway to manufacture adenine.
[00:28:24] Speaker A: Okay.
[00:28:25] Speaker B: And adenine is a really central, you know, molecule that's important for life because from adenine you can produce ADP and ATP, which is all about energy harnessing. You can produce nucleotides from it, you can produce amino acids from it. It's really important.
But the process to manufacture your own adenine is quite a sophisticated assembly line.
And for those able to see the slide, I think you'll be really impressed here with this particular graphic that came out of a biochemistry book shown at the bottom here. But what we have is an assembly line with 11 different enzymes, each of them shown graphically here. But you take the starting reagent, which is ribose 5 phosphate, and you have an enzyme to catalyze it to move on to the next step. And then an enzyme, another one, totally different enzyme, catalyzes it to the next step, another one to the next step, and it goes through 11 different catalyzed reactions to get almost to adenine at the bottom here. We're not quite at adenine, but we're. We're getting close at this point. So quite a sophisticated assembly line. And this particular yeast, this W3O3 yeast, they use it has one point mutation. This is well known in this yeast. One point mutation. That's actually.
It changed the codon to a stop codon.
[00:29:49] Speaker A: Okay.
[00:29:50] Speaker B: What that means is while this enzyme is being produced as a protein, it's being produced through the ribosome. There's an instruction to stop.
[00:29:58] Speaker A: Yep.
[00:29:59] Speaker B: And it just stops right there. So you get a little fraction of a protein, a fraction of an enzyme, which of course, doesn't work.
And as you can picture, an assembly line with 11 different steps. If in the middle of the assembly line, something stops, you know, what happens is you pile up parts behind it and you're stuck. You can't produce the product, but you also end up with a pile of, you know, the parts preceding it.
In this particular yeast, the pileup of parts behind this broken enzyme happens to be a slightly toxic product.
And so it's bad for the yeast to have this pileup of parts.
Now, if you, if you let these yeast evolve, it is possible that a random mutation will fix that one point mutation and you'll be back to producing adenine. And in this study, they did find out of 205 replicate populations, and there's, of course, billions of yeast in each of those populations.
There were. There were six of them that were lucky enough to have a point mutation to fix this, and they could actually produce their own adenine.
But what's much more common, what happens much more commonly, is that you get a point mutation earlier up the chain, up the assembly line, earlier in the assembly line, and that actually is beneficial because it then keeps you from producing this toxic intermediate.
[00:31:29] Speaker A: Okay? Yep.
[00:31:31] Speaker B: So now you have a situation where you have two point mutations and you're better off having those two point mutations.
But you can imagine now that if the yeast were lucky enough to fix one of those point mutations, if it fixed the early one in the assembly line, now it's going to go back to producing more toxin.
[00:31:54] Speaker A: Right.
[00:31:54] Speaker B: So it's actually worse off to fix that one.
If it fixes the second one down the line, that does nothing because you still can't produce your adenine. So you really have to fix both or nothing.
And that's where we get back to very similar to the study we did previously. Now we have two point mutations, and we're trying to see if yeast can fix those.
And they went through 10,000 generations of the yeast. And there's a quote here right out of the paper. It says, we do not observe any population that move from the lower Fitness genotype, meaning two mutations, to the higher fitness genotype. Meaning zero mutations able to produce adenine even after 10,000 generations of evolution.
So here we have an even more sophisticated assembly line with many thousands of nucleotides specifying how to make all of these 11 enzymes. And we see once again that if you have just two wrong letters out of all of that code, it's basically a lost.
You know, it's all broken, it's all lost.
[00:33:03] Speaker A: Right.
[00:33:05] Speaker B: Huh.
[00:33:06] Speaker A: And that's not to say that we're not suggesting here that any two changes in the code in this entire assembly line is going to do this, but these two specific ones that they were, that are required to be correct for it to function, those were the ones that they were focusing on, so. Good.
[00:33:24] Speaker B: Yeah. And you know, we're back to the question of saying then how can you teach me that evolution produced all of this from scratch? All these 11 enzymes, all the genes. Yeah, it produced those from scratch when I can give you at least 99.9% of a complete, accurate genome and it can't fix it.
[00:33:47] Speaker A: Right, Right. Yeah. And this again was measurable, it was repeatable. It sounds like they had over 200 different populations they looked at.
[00:33:55] Speaker B: Yep, exactly.
[00:33:59] Speaker A: Prospective in the sense that they set up the experiment and designed it in a way to focus on the particular issue that was at hand.
[00:34:07] Speaker B: Yep. And I don't think you could accuse these guys of any bias because this actually was like a side result of their study. It wasn't even something they were interested in.
It just turned out in the paper that's what they found.
[00:34:20] Speaker A: Okay, well, that kind of answers my, my prior question or my prior point, which is on the prior one, okay, do we, can we do this with another kingdom? Can we do this with another organism and see if we get the same result? And it looks like we do.
[00:34:33] Speaker B: It's fascinating and I think it gives us high confidence showing us what evolution is actually capable of.
Of course, what we're seeing is a very constrained process. You know, it, it can break things, it can tweak things, but it, but it's not going to be producing the kind of innovative biological pathway to produce adenine.
[00:34:53] Speaker A: Is this, is this surprising to you? Rob, if I told you I have a sophisticated functional system and I'm going to introduce copying errors into it and see what you know. Are we shocked that doesn't necessarily get better?
[00:35:10] Speaker B: I personally a little surprised that just too little two single point would shut the whole works down.
[00:35:16] Speaker A: Yeah, that's the surprising part, is that it's that bad. The situation has got to be that precise. Or it's a dire situation.
[00:35:24] Speaker B: Yeah. You know what I think, Eric? It all comes down to math and probability that, you know, if you have one point mutation, so there's like one out of a hundred million chance that you'll fix it. And these kind of single celled organisms produce tons and tons of organisms, so they can get over that. But if you have to have two, you know, simultaneous point mutations, now you're at 100 million times 100 million, which is like 10 to the 16th power.
And 10 to the 16th power is really hard number to reach from any kind of organism.
But certainly humans will never reach 10 to the 16th power population.
[00:36:05] Speaker A: Yeah. And again, this is very consistent with Mike Beh's work and what he's looked at in terms of the numbers. And like you say, it doesn't matter how much we want the theory to be true or in our heart of hearts, I mean it's, it's math, you
[00:36:19] Speaker B: know, it's just math.
[00:36:21] Speaker A: So, okay, show us the last one here.
[00:36:23] Speaker B: Left, last one. This one is an interesting paper because they wanted to know if random sequences of DNA could evolve to become something useful.
[00:36:35] Speaker A: Yep. Okay.
[00:36:36] Speaker B: That was the goal of the paper.
And so what they did in, in eco, we're back to E. Coli here. But they looked at, so E. Coli would rather not eat lactose as a sugar, it would rather eat glucose.
But if there is no glucose around, it'll turn on the production of enzymes to eat or metabolize lactose.
And the lactose genes, like everything in E. Coli, are organized in what's called an operon. So you have multiple genes that will be transcribed and translated when needed according to this promoter sequence, which will, you know, turn on the transcription.
So they wanted to see if they, if they just took out 103 nucleotides of DNA and then replace it with random sequences, totally random. And they had 40 different random sequences that they inserted in there to see what happened.
[00:37:31] Speaker A: So, and this was, sorry, just to be, just to be clear, this 103 nucleotides that they took out, that was previously a functional section?
[00:37:39] Speaker B: Yeah, that was the promoter.
[00:37:41] Speaker A: That was the promoter. Okay.
[00:37:42] Speaker B: And the promoter is there to say, okay, here's where we start under the right conditions. Here's where we start the transcription process. Right?
[00:37:49] Speaker A: Yep.
[00:37:50] Speaker B: Now the promoter in E. Coli is actually a pretty simple code that all you need, it's shown here, all you need are these six specific nucleotides in a sequence, a tataat.
And it's not quite as clear as you'd like it to be, but there's variations on that that are acceptable. Ideally, you have this six sequences, then you have a spacer of 17 other nucleotides and then another six nucleotides in sequence to be a successful spacer.
[00:38:23] Speaker A: Okay. And that's somewhat. That's somewhere in the 103.
[00:38:27] Speaker B: That's right. Anywhere in there you could get if you have this sequence.
[00:38:30] Speaker A: And just curious, I know this isn't part of the. What you're talking about, but just curious, is the 103 length important here? Is the.
[00:38:38] Speaker B: No. I'm not sure why they picked 103, but clearly covered it.
[00:38:42] Speaker A: Sorry to distract you there. Go ahead.
[00:38:44] Speaker B: Yeah. Not sure why it was 103, but again, the system is not hyper specific about having these exact codes. There's a lot of variations that it will accept.
So anyway, they plopped in these random sequences and then they grew the mutated E. Coli in a solution that had a little bit of glycerol and. And then it had to go to lactose. If you want to keep growing, you better be able to eat the lactose.
And what they found is that 95% of these random sequences were able to evolve, to produce and to be able to metabolize the lactose. So that. That's a lot of success. 95%.
But what's interesting is why or how they evolved in order to. To get that job done.
So what they found is that in 10, 10% of the 40 sequences, or in four sequences, they found that the sequence already was an acceptable promoter. So just random generation, it's already good. You got what you need.
So that shows kind of how simple of a task this is. If. If this randomness can get it done, then they found another, a group of 57% of the sequences.
All they needed was one point mutation.
[00:40:03] Speaker A: Okay.
[00:40:03] Speaker B: So a single point mutation, which we've already seen in the other studies, that can happen. And now you have a successful promoter to metabolize lactose.
[00:40:12] Speaker A: Right. So that's 2/3 of the situation.
[00:40:15] Speaker B: Yeah, we're already 2/3 through the PI. So now you would think, okay, what's next is having two mutations. Because we did zero, we did one, now next thing would be two.
But I think we learned from the two previous studies that that's not easy.
[00:40:28] Speaker A: Yeah.
[00:40:29] Speaker B: So in fact, what they found consistent with the other two studies is, you know, 5% of the sequences, they never found a solution, couldn't produce enzymes to metabolize lactose.
Then in another 12%, what happened is instead of Having two mutations, it was actually a deletion that occurred in the termination sequence. So different operons are separated from each other by termination sequences.
[00:40:59] Speaker A: Yeah.
[00:41:00] Speaker B: Or else you just keep translate transcribing forever and ever, and that's not a good thing.
So they're separated by boundaries, and if you delete that boundary, then the previous operon will also transcribe the next one, and you end up producing the lactose enzymes.
[00:41:18] Speaker A: Just keep on transcribing.
[00:41:19] Speaker B: Yep, that did it. So that was 12% of the sequences.
And then in the final group, which is 15% of the random sequences, what happened was somewhere a promoter, a different promoter that was upstream was kind of copied and pasted into this region here. So they got a promoter somehow, but they didn't get it at all through two random mutations occurring that never happened.
[00:41:46] Speaker A: Yeah. Can I ask you about the upstream promoter? So I don't know if they. If they had this much detail in the study or if you remember, but did they run the numbers on that particular situation to understand what was going on in terms of the randomness? So my understanding is, first of all, promoter capture is one of the most common types of changes in bacteria.
And it's not necessarily clear that it's always a random process.
You're grabbing a promoter, you're not just grabbing a nucleotide, you're grabbing a sequence that's a set of specific things, and you're taking it from here and saying, okay, now insert it over here.
There's something about that process that we're still researching and understanding, and it doesn't seem to be completely random, particularly in these bacteria.
[00:42:34] Speaker B: Yeah, I totally agree. And the same is true of duplicating genes.
[00:42:38] Speaker A: Right? Exactly.
[00:42:39] Speaker B: Very common thing. And deleted extra copies of genes are a very common thing.
That doesn't seem like a totally random process. Or you might.
[00:42:47] Speaker A: Yeah, it can be otherwise.
[00:42:49] Speaker B: Eugene. Or three quarters.
But it's interesting how often it happens. In this paper, they didn't go into great detail about these relocated upstream promoters.
[00:43:00] Speaker A: Okay.
[00:43:01] Speaker B: They didn't go into it, but I. But I think you bring up a really good point.
[00:43:04] Speaker A: One of the. There's a group that's been working on this a little bit that I've got some connection with, and I think you do too. But there's some additional research now being done to look at some of these types of stochastic approaches that these large populations can take to things like deleting upstream terminations or capturing promoters, copying full genes, and it's gonna be interesting to see where that shakes out, but it does not seem to be a fully, you know, sort of random mutation.
[00:43:33] Speaker B: Yeah, fascinating.
[00:43:35] Speaker A: Yeah, good.
[00:43:36] Speaker B: So to wrap this up, we have three studies that meet the criteria for high confidence science that we've been talking about. So I think you can, you can trust these results very clearly showing us that the process of evolution, you know, random mutations and natural selection is, is very constrained in what it can accomplish. And very quick to kind of give up on complicated systems, biochemical pathways.
[00:44:06] Speaker A: Right, right. And again, you know, I've shouted out to Mike a couple of times. I know you did as well, but just sort of supports the work that he did a number of years ago in terms of Darwin. Duvall's was the title of his book, and some of the work that he did showing the challenge of getting beyond anything more than a couple of point mutations.
And you end up, like you say, it's math. You end up requiring massive population sizes, massive numbers, and even in those cases, you very quickly run out of resources to be able to do anything that's, that's moving in the right direction. And you run into the problem that a short term gain is much easier to obtain by breaking something or deleting something or blunting something.
[00:44:50] Speaker B: It's simply more probable. And so that's the path that it takes.
[00:44:54] Speaker A: Yeah. Yeah. Okay. All right. So if you had your way, and you were writing a textbook today, Rob, and you were taking your son's 9th grade biology book and reworking it a little bit, what kinds of evidence would you like to see shown more discussed, more acknowledged, more in terms of high confidence evidence versus some of the low confidence evidence that they're talking about.
[00:45:19] Speaker B: Yeah, good question. I mean, these three studies that we've gone into in detail, and they're covering different kingdoms of life, and they're covering it with very high confidence. And I think it's the kind of thing that the textbooks ought to be considering.
I would like to think that students are being taught that high confidence evidence should be trusted and relied upon more than low confidence evidence, and it should be prioritized and the interpretation should come out of that.
[00:45:51] Speaker A: Right, right. And the good thing about these last three studies that I, like you talked earlier about taking your statement or your claim and rephrasing it as a question.
So these three studies are asking maybe the second one a little bit indirectly, but certainly the other two and the third one, even in an indirect way, what can evolution actually accomplish? What can random mutations actually accomplish? That's a question. Rather than saying, well, we know that it can accomplish these things. Therefore, we're going to point to similarities or we're going to point to the fossil record and we'll just make our claim based on that. They're actually asking the question.
And when you ask the question and set up the study in a repeatable, measurable, prospective way where you control the experiments, the evidence is not very impressive for what evolution can do.
Pretty sobering.
[00:46:45] Speaker B: It is very sobering. I'd like to add two final points.
[00:46:49] Speaker A: Sure.
[00:46:49] Speaker B: Which is that if a single mutation confers a benefit, that I think is a scenario where maybe you can make more progress than what we're seeing here. Although we didn't see it in the E. Coli tryptophan experiment necessarily. But if every successive mutation causes a benefit, then I think more progress could be made than this.
And the second point is that I've had some criticism about this where they'll say, hey, neutral evolution could come in and fix all of this. You could just have neutral evolution. But there was nothing about these experiments that blocked out the possibility of neutral mutations happening. I mean, you just let mutation, you just let evolution do its job. And as we've said through analogy, it kind of tripped and fell at the one yard line without being inhibited from doing its thing. It did its thing and it fell.
[00:47:46] Speaker A: That's a really important point that you made. There's nothing in any of these that's preventing the neutral mutations. In fact, we have to assume that if they existed, they were there among these thousands of generations and billions upon billions of organisms. So that really doesn't add anything. And the great thing about these three studies is we don't have to rely on what ifs or questions or theoretical statements. We have data and we can actually look at the data. And the data is pretty clear what it's telling us.
[00:48:19] Speaker B: Absolutely.
So let's treat it with appropriate appreciation for being high confidence.
[00:48:25] Speaker A: There you go. There you go. All right, well, we'll push to get these in the textbooks instead of some of the lower confidence stuff. This has been fantastic. Rob, any final words as we wrap up here?
[00:48:38] Speaker B: No, just thanks for your time. I appreciate it and I'd be happy to people know how to contact me. Be happy to answer any questions. Thank you.
[00:48:46] Speaker A: Okay. Appreciate it. Absolutely. And thank you so much for spending time with us and for sharing this fascinating information. Really appreciate it. Thank you, Eric for ID the Future. I'm Eric Anderson. Thanks for joining us today.
[00:48:59] Speaker B: Visit
[email protected] and intelligentdesign.org this program is
[00:49:05] Speaker A: copyright discovery Institute and recorded by its
[00:49:08] Speaker B: center for Science and Culture.
