That’s only one way of getting a DFE, and it’s really only applicable to microbes. It’s definitely not the way that Racimo and Schraiber obtained their genome-wide DFE for humans.
You’re confused about a couple of things here. First, epistasis does not represent a change to the DFE as you seem to think it does. It represents a dynamic of interaction between mutations existing concurrently in an individual. Acknowledging the presence of epistasis in no way entails acknowledging that DFEs are not stable over time.
Second, diminishing returns are not what you need to escape Kondrashov’s Paradox and Genetic Entropy. You need the exact opposite. You need the DFE to move from the state we have evidence for now (which is causing fitness decline) to a state where fitness decline is not only arrested but moves in the opposite direction. Or you need epistasis to reverse the effects of the DFE – so that would be some kind of synergistic epistasis among beneficials coupled with antagonistic epistasis among deleterious. This, naturally, is the opposite of what we observe in the real world: deleterious in combination become somewhat more deleterious, and beneficial in combination become less beneficial.
Although it’s worth noting that Dr. Masel seems to have ruled out epistasis altogether if you read her website. She basically discounts the idea that epistasis is much of a factor at all.
I did simulate synergistic epistasis among deleterious mutations at both a light level and a strong level, since this was actually one of Kondrashov’s proposed rescuing devices. Sanford himself had previously modeled the same. Both his simulation and mine confirmed that synergistic epistasis among deleterious mutations only speeds the decline of fitness.
So let me ask you again: do you have any proposed mechanism for how you are going to cause the DFE which we have published evidence for to morph into a different DFE that can not only stop the decline but reverse the damage? Or some kind of epistasis effect that would result in the same?
Dr Sanford noted that defining fitness in terms only of reproduction is a circular argument. He suggested instead that fitness be defined in terms of real traits and abilities like intelligence or strength or longevity.15 In other words, does the organism appear to be getting healthier over time, or weaker? Genetic entropy is not really directly about reproduction—it is about the decline of information in the genome. We should expect that as our genes are damaged, various physical traits would begin to decline as a result of this damage, and this decline will at first be more noticeable than any possible reduction of the ability to reproduce (this is especially true in humans, since we have advanced modern medicine to help us).
Conclusion
Only when we insist that the terms of the debate be fair and accurate will we have any chance to clearly communicate the truth of creation and the bankruptcy of Darwinism to the world at large. As we have undertaken to demonstrate here, evolutionists have been guilty of hiding behind the misleading use of the term ‘fitness’; it is now time for honest scientists to adopt a more realistic, objective look at the parameters of life. When one does this, the picture is bleak. We are all dying a slow death of genetic decay.
Intelligence, strength, and lifespan, are traits, which appears to be inconsistent with your rejection of phenotypical expression as having a measurable objective role.
The bigger problem is that the only direct proof of breeding success is in the pudding. If there is a way that genes may continue, be they deleterious, neutral, or beneficial, without reproduction, please do say. Otherwise, no matter how strong or clever, without offspring, those genes are wiped from existence.
The Understanding Evolution website of the UC Berkeley Museum of Paleontology supports the straightforward and conventional discipline understanding of fitness I provided earlier.
Evolutionary biologists use the word fitness to describe how good a particular genotype is at leaving offspring in the next generation relative to other genotypes. So if brown beetles consistently leave more offspring than green beetles because of their color, you’d say that the brown beetles had a higher fitness. In evolution, fitness is about success at surviving and reproducing, not about exercise and strength.
…A genotype’s fitness depends on the environment in which the organism lives…
…Fitness is a handy concept because it lumps everything that matters to natural selection (survival, mate-finding, reproduction) into one idea. The fittest individual is not necessarily the strongest, fastest, or biggest. A genotype’s fitness includes its ability to survive, find a mate, produce offspring — and ultimately leave its genes in the next generation.
So Paul it occurs to me you need a visual aid to understand how diminishing returns epistasis would change the DFE. I made this simple animated figure to help you:
It is intended only as a visual aid to understand how the phenomenon known as diminishing-returns epistasis would result in shifting the DFE towards a higher proportion of beneficial mutations as organismal fitness declines.
So, to explain what we see, as W rises (the population climbs a fitness peak, say), beneficial mutations become more rare and have smaller magnitude of effect. This is represented in this animation by the red area right of “0” on the X-axis, which are the beneficial mutations in this DFE, becoming smaller.
If the population could truly get all the way to the absolute peak, there would be no more beneficial mutations and therefore the DFE would shift entirely to having deleterious mutations.
But a corollary is that, if you move down from the fitness peak, the reverse happens. As W diminishes, the red area right of “0” on the X-axis, which are the beneficial mutations in this DFE, grows larger.
Hopefully this should clear up any remaining confusions you have.
This also applies to explaining the magnitudes of differences, from direct measurements of heterozygosity and polymorphism, between groups of humans. Even if real, no observations of different mutation rates between groups come close to their magnitude.
Now, if you had real faith (both secular and religious) in a young earth, you’d be voraciously chasing these data down because you have complete confidence that they will confirm your position. Your lack of interest indicates the opposite.
But over transcription and translation is a loss of information, or, more precisely, a loss of meta information, which refers to information about how biological information is used. IOW, a higher-order layer that describes, regulates, or contextualizes the primary biological data without itself being that primary data.
What I wrote is absolutely correct and applies to all species you want to record some DFE for.
Certainly there are different ways of obtaining and estimating fitness for different mutations (you can’t measure growth rates in humans to obtain a fitness-effect for a single mutation of course), but since I didn’t specify how exactly you “record the fitness effect”, you have no basis for levying a complaint.
You seem able to only ever give uncharitable and negative readings of other people’s words. It’s amazing.
I am not confused about anything since I don’t think the concept of epistasis (which in isolation just means that mutations have contextual effects) represents DFE change.
The article I link is speaking about a specific type of epistasis called diminishing-returns epistasis. Do you understand?
Key hints for you, Paul, are the words ‘diminishing’ and ‘returns’ occurring immediately prior (to the left of) to the word ‘espistasis’ in the sentence. Since we read from left to right.
Yes, there are many forms of epistasis. Synergistic, antagonistic, sign epistasis, additive, multiplicative, negative, positive, local, global, and so on and so forth. They all mean different things.
So it’s you who seems confused by me linking a paper using the term diminishing-returns epistasis when offering a mechanism to explain why beneneficial mutations become both more prevalent, and have larger magnitude of effect, in less fit backgrounds.
I hope this helps but will be happy to explain it to you if you still have trouble getting up to speed.
You are so confused it boggles the mind. Diminishing returns epistasis is the phenomenon where adaptation slows down as you approach a fitness peak. As the population adapts (mean fitness rises) adaptation (the rate of fitness increase) slows down.
The explanation for why this (rate of fitness increase slows down) happens is that beneficial mutations become both increasingly rare (the pool of available beneficial mutations shrinks), and their magnitude of effect becomes smaller too.
The direct implication of this is that, when you move AWAY from the fitness peak, the OPPOSITE happens. As fitness gets lower (which is what happens to the population in your simulation), if we were to include the observed phenomenon of diminishing-returns epistasis then MORE beneficial mutations become available, and the SAME beneficial mutation in a less fit background has a greater magnitude of effect.
That means, in the context of Kondrashov’s paradox, that as population fitness declines (deleterious mutations accumulate in all individuals) we move away from the fitness peak. And because we move away from the fitness peak, fitness gets lower, and because fitness gets lower now MORE benefical mutations become available, and their average magnitude of effect becomes larger.
So yes, in fact, diminishing returns epistasis would constitute a phenomenon acting to suppress the effect of accumulating mutational load, by increasing the beneficial effects of available beneficial mutations, and making beneficial mutations more frequent. For that reason it would, in fact, shift the mean mutational effect of the DFE by growing the beneficial portion. The beneficial mutations would become more numerous, and their magnitude of effect would increase.
It’s really sad I have to explain this to you because you apparently don’t understand that the there are many different classes of epistasis. Or you’re playing dumb or something. It’s truly strange.
This is just a fantasy on your part. You can’t “rule out epistasis” since there are many different forms, and you’re just hilariously and painfully wrong.
Read the paper again. And if you still don’t understand how this is in fact an answer to your question, read it a third time.
Granted that the broader genome is involved in the timing and degree of expression of genes. But the real problems are the entire creationist insistence that the genetic code is sacred information being itself meta, and further that they cannot or will not provide a definition.
More specifically, whether virus, bacteria, or animal, rough existence has tuned trial and error feedback to template proteins involved in the development and maintenance of an organism over its lifespan, and this has been much observed. We see the preserved and ongoing imprint of that ceaseless cycle, no more or no less. In that limited sense the very general word information may be applied, but that specific usage cannot be abused to serve as a Trojan Horse for any further meaning in the dictionary, and particularly for any suggestions of intelligence or idealized perfection.
Is that what you are going to go with when you get around to the choir? I went on the evolutionist’s expert forum, and I gave peer reviewed sources and all they did was quote wikipedia and AI slop?
Peer reviewed papers I cited and you ignored or just brushed off:
Meanwhile, you have referenced non reviewed books such as some unknown volume on live dinosaur sightings, let alone Sanford’s popularized Genetic Entropy.
Now consider what Paul says here in response to me pointing to diminishing-returns epistasis (which causes beneficial mutations to gain larger effect size when fitness declines) as a solution to Kondrashov’s paradox:
Paul also completely ignored that I cited a paper that directly references the phenomenon as a potential solution to the mutation load problem in species under threat of extinction:
Importantly, the type of global epistasis discussed in that paper is diminishing-returns epistasis. It appears Paul was not aware. Apparently he didn’t even read the paper, since it he hasn’t clicked on the link. If he had read reference 44 therein, he would have then had this fact become clear to him.
But he wouldn’t like that paper for other reasons too, such as the failure of the authors to find unconditionally beneficial or deleterious mutations, once again supporting what we have ALL been trying to explain to Paul ever since he first graced this forum, that fitness relates to the environment.
I wrote her and asked (and linked this thread), and that doesn’t seem to be her stance at all. I asked her permission to quote what she wrote here:
Kondrashov believed the solution was epistasis, specifically what I would call “negative epistasis” rather than “diminishing returns epistasis”. The latter term is usually used for a series of beneficial alleles, the former term is more general, including the case of diminishing returns epistasis but also including a series of deleterious alleles or a mixture of beneficial and deleterious. When low fitness makes compensatory beneficial mutations more frequent, that is described by a different term, namely “sign epistasis”, meaning that the same mutation that is deleterious in one context is beneficial in another. It’s a real thing, with perhaps the clearest evidence being this paper: https://doi.org/10.1126/science.adn0753. My lab has not “ruled out epistasis altogether”. Measurements put average negative epistasis among deleterious mutations as being of very small magnitude and I don’t think would be enough in a world where each deleterious mutation needed to be compensated by one beneficial mutation. But sign epistasis can certainly be an important part of the solution, as could negative epistasis (=diminishing returns) among beneficial mutations. The solution in our preprint was that a smaller number of large effect beneficial mutations compensate for many smaller effect deleterious mutations. Mechanisms that increase the number and size of beneficial mutations, as you propose, would interact synergistically with what we proposed to produce an even stronger solution.
I’ve been skimming over your posts in my email over this past week as I’ve been on vacation.
Your fictionalized series of DFEs were wrong, as you’ve got the wrong thing on the X axis. Every DFE I’ve ever seen in print has s, or selection coefficient, on the x axis. This is the effect on fitness for the individual mutation in isolation. It has nothing to do with modeling some theoretical combined effect over time as a mutation load grows.
With the possible exception of this “sign epistasis” concept, epistasis has nothing to do with altering the relative frequencies of beneficial and deleterious mutations on the DFE. It’s about how the mutations that happen to coincide in the same individual will act in combination with each other.
You’ll note that Dr. Masel is not claiming to have actually solved Kondrashov’s paradox, but only that she has been working on some proposed solutions. I will agree that “sign epistasis” (that’s the first time I’ve heard that term) could happen with microbes in a lab, but overall it is not likely to be a major factor in LMEs. Why? Because as I’ve explained numerous times and you’ve preferred to ignore, mutations are overwhelmingly deleterious regardless of environment. It’s not the environment that makes them so, it’s entropy. It’s the fact that they are random changes to functional complexity. That’s what Gerrish explained.
Sign epistasis is apparently the technical term for the claim that as the fitness of the population declines, something (?) causes the proportion of beneficial mutations on the DFE to spike upwards and come to the rescue. I wonder, why didn’t that happen and save the wooly mammoth? If sign epistasis were a real functioning rescue mechanism to prevent mutational meltdown, then we should never observe it in the first place for anything (but we do).
Something like this was actually the first proposal we responded to here about 6 years ago. It’s worth noting that my more recent research and use of SLiM has updated my understanding of mutation-drift equilibrium since the time that article was written. I now believe, based on my own SLiM simulation, that mutation-drift equilibrium will be reached in humans long before extinction, and it’s not an ever-increasing segregating load that causes the fitness decline, but rather the ever-increasing mean deleterious allele frequency (Ohta’s Ratchet). I explained this in detail in the Hancock debate.
This means that, past a certain point, you do not have an ever-greater number of mutations creating an ever-greater number of opportunities for back mutations, as was claimed by some. As we remarked previously, back mutations are inherently less likely than the forward mutation, and since fixation is not happening on the timescales we’re operating under, this is not much of a target for them to hit in the first place. Fixation for neutral alleles takes on average 4Ne generations, and that would equate to about 800,000 years in humans (that’s when the very first substitutions would start trickling in). Yet extinction is expected in a tiny fraction of that number – about 40,000 years estimated for my simulation.
Looking at the article that Dr. Masel cited for sign epistasis, you may notice one very important statement buried late in the paper in the discussion section:
One potential shortcoming of this theory is that it ignores the biological architecture of organisms, which could impose some nontrivial structure on epistasis (51).
Indeed! A few experiments with yeast that show sign epistasis is a far cry from showing that large multicellular eukaryotes are going to be somehow, as if by magic, rescued from fitness decline. The underlying function of biological information, and the ever-increasing layers of complexity as you move from simple life to complex, must be taken into account.
Regarding the presence of epistasis in general:
I did include synergistic epistasis in my simulation (so since this is your definition of “not fixed” for a DFE, that means you’re wrong to claim my simulation was fixed). The more SE you add, the faster the fitness decline happens. Diminishing returns epistasis would not help you at all, it means that the beneficial mutations work against each other in combination.
When Dr. Masel says “a smaller number of large effect beneficial mutations” she is just repeating the same rescuing device that Kimura appealed to in 1979, and which has been rebutted by Dr. Sanford himself in his original book. It’s entirely unrealistic for LMEs, and it’s also not at all observed in LMEs. When you use realistic numbers as I did, fitness decline always happens.
If you’d like to discuss this some more, I strongly suggest you visit the upcoming post-debate livestream at SFT:
Researchers have isolated or developed virus strains with much lower rates of mutation than their wild counterparts, in order to study how beneficial or deleterious the propensity to mutate is in and of itself. While excessive mutation would inhibit essential functions, does this mean that the less mutation, the better? A great deal of broad based research has delivered an answer, and the verdict is no. Mutation is necessary for viral survival.
Using previously characterized CHIKV fidelity variants, we addressed whether CHIKV population diversity influences the severity of arthritis and host antibody response in an arthritic mouse model. Our findings show that CHIKV populations with greater genetic diversity can cause more severe disease and stimulate antibody responses with reduced neutralization of low-diversity virus populations in vitro.
Using chikungunya virus (CHIKV), we describe a unique arbovirus fidelity variant with a single C483Y amino acid change in the nsP4 RdRp that increases replication fidelity and generates populations with reduced genetic diversity. In mosquitoes, high fidelity CHIKV presents lower infection and dissemination titers than wild type. In newborn mice, high fidelity CHIKV produces truncated viremias and lower organ titers. These results indicate that increased replication fidelity and reduced genetic diversity negatively impact arbovirus fitness in invertebrate and vertebrate hosts.
The results indicated that these high-fidelity variants exhibited an attenuated pathogenic phenotype in vivo and offer promise as a live attenuated EV71 vaccine.
It has become increasingly clear that increases in fidelity always come with a fitness cost, as higher fidelity mutants replicated more slowly in cell lines.
the 3D-G64S high-fidelity phenotype reduced viral fitness under a defined selective pressure, making it likely that the reduced spread in murine tissue could be caused by the increased fidelity of the viral polymerase.
A RNA virus is an infectious agent requiring one essential resource actively defended by host organisms, so its environmental and lifecycle challenges are distinct from larger eukaryotes, but a general and self-evident principle is supported by this research. Fidelity is a vulnerability which is dependent upon the benign continuity of a static environment, and variation is essential for adaptation. We have survived a hundred times over due to mutation.
Still cannot or will not answer the simple question, is a mutation resulting in phenotype color change deleterious, neutral, or beneficial? I’m sorry, is that inconvenient to your rhetoric?
That is not reality, and you are full of creationist conference gibberish.
I would remind you that the inaugural published paper which is often referenced in discussion of genetic entropy, and is prominent in Sanford’s popularization, dealt with influenza virus. Now that is has been shown that GE is irrelevant in microbiology, you want to wave away the demonstrated principles at work. That is what I would expect from an apologist who has little real interest in nature.
Do you know what functional complexity is Paul? They are called traits, those things you deny have an objective existence.
Who made you qualified to judge what is realistic or unrealistic in terms of numbers or model? Empirical observation is definitive that you are wrong.
Then you simply misread the figure. It’s selection coefficient on the X axis. I admit I could have done a better job making it clear what the different parts are, for example the W(fitness)=y line should have been outside the figure as it merely conveys the idea that the distribution shifts as a function of fitness (as you can see the DFE shifts between three different frames, each with it’s own label).
But honestly you should have been able to work this out yourself, both because of my explanations, because you’ve seen DFEs before, because I’ve shown you this exact figure before about 6 years ago and explained it then too, and since it says fitness effect below the x axis any way.
As usual this is just you being intentionally obtuse for the sake of appearance. I don’t believe for a second you didn’t know or understand what you were looking at. But you know what? Here’s an updated figure just for you Paul:
Look, all of this is basically captured by the idea that the population climbs a fitness peak and adapts to it’s local environment.
Wrong. Diminishing returns epistasis is just another example of how the DFE can shift because the same mutation, still beneficial, has a larger magnitude of effect in a less fit background, and a smaller magnitude of effect in a more fit background. Dude just read the damned papers I link you. You haven’t clicked or read a single one.
Epistasis is about that too, sure.
I don’t know what you think the word “solution” means but I have some news for you then.
Because then you’d have a problem of course.
And they can still be. You can still get a fitness increase with a DFE with mostly deleterious mutations. Because of the massive differences in the probability of fixation, a relatively minor increase in the frequency and magnitude of beneficial mutations can counteract the preponderance of deleterious mutations, even when they are still strictly speaking more likely to occur.
So you can still have a DFE with a mean fitness significantly on the deleterious side, but where fitness goes up as the population evolves.
That is, after all, what experiments show happen even when the measured DFEs change, they are still skewed towards mutations of small effect and mostly deleterious, but the small changes in magnitude and frequency to both the deleterious and beneficial side enables fitness increase and declining rates of it.
To say I have ignored this point is just astonishing when I have explained this exact point to you before numerous times.
Neither you nor Sanford have done any work to show that the shape of the DFE is due to entropy, nor would such work imply that the shape of the DFE can’t change as a function of fitness, or that selection can’t work to increase fitness against a preponderance of deleterious mutations.
If it did we would already know such work was false, since it would then conflict with the direct empirical observation of what happens in experiments, where fitness goes up.
Read the papers I link Paul. Read them. Click the links and read them.
No, sign epistasis is when the effect of a mutation reverses (deleterious to beneficial, beneficial to deleterious either in a different environment, or in a different genetic background.)
An example of the former would be when a mutation in a membrane transporter might reduce it’s activity under normal conditions, making it deleterious, but that same mutation completely blocks the uptake of an antibiotic, giving the cell strong resistance, making it beneficial when antibiotics are present.
For the latter (different genetic background), an example could be the formation of a stabilizing disulfide bridge in a protein. The first mutation to cysteine might be deleterious, but if another residue elsewhere has already mutated to cysteine then the disulfide bridge can form and both mutations are now beneficial.
To be sure, sign epistasis can contribute to decelerating adaptation due to diminishing-returns epistasis (by making some previously deleterious mutations become beneficial when fitness declines), but they are not the only cause of it.
Dimishing-returns epistasis among beneficial mutations specifically is the phenomenon where the SAME beneficial mutation gains a LARGER magnitude of effect in a less fit genetic background (or in a more stressful/poorer-quality environment). That’s not sign epistasis (the effect of the mutation doesn’t reverse). As for how that could happen, read the papers I link Paul. Read for comprehension.
There is also diminishing-returns epistasis among deleterious mutations where the magnitude of effect of deleterious mutations diminishes as fitness declines. Interestingly the magnitude of effect of deleterious mutations (the same mutation in a less fit background) diminishes as fitness declines. Experiments have demonstrated that both phenomena (diminishing returns among beneficial mutations as populations adapt, and diminishing returns among deleterious mutations as populations decline) co-occur.
There’s even the third contributing factor to why the you get decelerating rates of adaptation due to diminishing-returns epistasis, which is simply that if deleterious mutations accumulate, more and more targets for reversal open up. Yes, new deleterious mutations can still remain predominant (because among other things, sign epistasis will also open up more deleterious mutations).
This is classic black and white thinking. It just doesn’t follow that because diminishing returns epistasis CAN counteract accumulating mutational load that it will always do so under all circumstances. You seem to be saying something like if mechanism A can help prevet situation S, then we should not ever observe situation S under any circumstances. Like, that’s just obviously fallacious reasoning.
Besides, the article doesn’t even claim the Wolly mammoth went extinct due to mutational load and simply leaves that an open question.
After all, it says:
It is unlikely that any single mutation doomed the Wrangel mammoths—it still isn’t clear exactly what led to their eventual extinction.
Could it have been a contributing cause? Sure.
Nope. Nothing there relevant to my points about diminishing returns epistasis, changing DFEs, or beneficial mutations gaining larger magnitude of effect when fitness declines (or, also importantly, deleterious mutations having smaller effects).
You have no mechanism that explains why it wouldn’t also happen to “large” multicellular eukaryotes. Why isn’t the multicellular fungus large enough? How much larger does it have to be before this magical fantasy-DFE-fixing-force starts to take effect?
This is inept to a degree that is difficult to overstate. Synergistic epistasis is when two mutations have greater than additive effects (so two deleterious mutations become even more strongly deleterious together, and two beneficial mutations become more strongly beneficial together ). That wouldn’t change the DFE, since the same effect would operate on both deleterious and beneficial mutations assuming you kept both sides with equal synergistic values.
So you tried a run with synergistic epistasis among deleterious mutations only (they could combine in a way that made their combined effects worse than additive, but from the name EPS_DEL synergistic epistasis it appears it only affected deleterious mutations, making them worse).
Furthermore, you have labeled “EPS_BEN” as “diminishing returns” and set it to 0.0 and curiously call that the “best case for evolution”. It’s not at all clear that EPS_BEN is actually diminishing returns epistasis (as opposed to simply the synergistic effect on beneficials that EPS_DEL was for deleterious), and if it was in fact diminishing-returns epistasis (the one where the same beneficial mutation gains a larger magnitude of effect in a less fit background) it’s just flat out obvious setting it to 0.0 is not, in fact, the best case for evolution.
I’m starting to have questions about your simulation Paul.
Also, just to reiterate, synergistic epistasis isn’t diminishing-returns epistasis. Like, can you learn what the terms actually mean Paul?
Looks like we hit the bare assertion fase. When all else fails, just start blatantly asserting what you’re being asked to prove.
