And yet he presents no argument for why the bulk of such RNAs ought to be functional, nor does he attempt to counter any of the arguments that they shouldn’t be. He says that a few of them have known functions and infers from that that all of them must be, just as I said. The main argument for junk DNA is that it isn’t conserved within or between species. And in the case of these lncRNAs, most of them exist in extremely low concentrations, on the order of one per cell.
Not only that, but strongly deleterious. I believe the simulation depends on them being nearly neutral, but comparison between species shows strong purifying selection at most sites.
I used the DFE published by Eyre-Walker and Woolfit. If you have a problem with their work, please 1) say what your problem is with their work and 2) take it up with them.
Are you talking about mutation frequency or the percentage of possible mutations? Remember that transitions are much more common than transversions, C-T most common of all. Also, you should know that there are two amino acids for which some first position mutations are silent in the universal code.
That sounds like a number that doesn’t take into account the frequencies of mutation types.
Some of them are strictly neutral, some are deleterious to such a slight degree that they would be effectively neutral even in the largest populations, and the occasional one is significantly deleterious. But the short approximation is that they’re neutral.
1 Like
Bro, Sanjuan et al. is single-nucleotide substitutions in an RNA virus. They’re not just nonsynonymous mutations in coding regions here.
Besides, the proportion of beneficial mutations in that paper is over 4%. And yet you just bullsh1tted me that the ration is supposed to be 1000:1 deleterious:beneficial for amino-acid changing mutations. A ratio you’re citing Sanjuan 2004 at me to support. Have you got a bit confused here perhaps?
Remember, we were discussing this:
So you used wildly different numbers from all over the place.
Also, the beneficial effect-size for beneficials in Sanjuan et al. 2004, which apparently you now think is appropriate, was between 1% and 5%. These are hugely beneficial mutations. I’m going to go ahead and suggest you did not use those values in your simulation.
It’s rather strange because on the one hand you earlier complained about using DFEs from ostensibly inappropriately simple species like bacteria, or heck, even multicellular eukaryotes like Drosophila and A. thaliana, but here you’re citing Sanjuan et al. 2004 (from an RNA virus) at me to support a number you don’t even use.
Further, “consensus view”? I feel like the honesty of your position has been suffering somewhat of an ongoing decline in this discussion. Entropy of a kind shall we say.
But hey let me be charitable here and say I think you got confused and forgot what you were supposedly trying to model.
1 Like
This may be an underappreciated aspect of the Genetic Entropy argument, that the amount of unselectable deleterious mutation is book ended between strictly neutral mutations and moderate and strongly deleterious. Anything above effectively neutral will not become fixed given a sufficient population.
The Omim database of human genetic diseases lists thousands of known Mendelian disorders associated with mutations to over 16,000 genes, so most coding DNA. These are not near neutral, they are disadvantageous and often devastating to the phenotype. Though definitely deleterious they do not contribute to purported Genetic Entropy.
That leaves deleterious mutations in DNA that matters, yet matters little; that is blind to differential reproductive success, and proceeds to complete fixation, all while registering cumulative effect. It is not clear what the operator between individual slightly deleterious mutations is; if a mutation inactivates a gene with a given fitness effect, does a subsequent mutation to that gene register the same degree? Is it straight up addition, and when a certain tally is reached it is game over? On the path to extinction we are supposedly on, what actually does the killing? After we whisper rosebud, what condition is written into the pathology report?
In the interest of apologetics, @UncensoredPilgrims seeks to magnify the scope of deleterious mutation as high as it can go, involving essentially all of the genome, but at the same time to trivialize any immediate phenological disadvantage.
1 Like
I don’t have access to that paper, but if we go by the title that number may not mean what you think it does. Is it perhaps a Ka/Ks ratio? And if so, did you happen to look at what that means?
1 Like
You seem to have a basic confusion between coding and non-coding DNA. Note that Eyre-Walker & Keightley (I have no access to the full paper, and I’m not clear on who Woolfit is) have 2 DFE, one for coding and one for non-coding DNA. Which did you use, and which sort of DNA were you representing? Again, you are very unclear about what you’re doing.
Delete
1 Like
You need to pay closer attention to what I write. I used the 1000:1 ratio which is cited in the Sanjuan paper as the consensus view. It’s not the observed ratio in their RNA virus experiment, which for all the reasons I already mentioned in the debate, is not relevant to GE in LMEs.
I would say from your side there has never been much honesty to begin with. First you accuse me of simply making up numbers. Then I show you that not only did I not make them up, they actually represent the consensus. And this is the best response you can come up with.
There’s no point in being charitable. It’s obvious this has run its course and there’s no point in continuing to beat a dead horse.
It means exactly what I said, and in fact this number can be easily obtained through a simple Google search. You might try it before pointlessly arguing.
My actual simulation - the one I believe is most accurate - used Racimo and Schraiber’s genome-wide DFE for humans.
The latest simulation I did off the cuff was in response to Felsenstein’s suggestion that we can simply ignore the vast majority of mutations since they allegedly happen in junk DNA.
To test that claim, I used the mutation rate he suggested (7.2 per generation), and I reduced the entire genome down to just the 1% or so which would be non-synonymous coding mutations. Then I used Eyre-Walker and Woolfit’s DFE, the parameters for which are as follows: gamma distribution, shape 0.23 and mean 0.043. Their mean is orders of magnitude higher than Racimo and Schriber’s, since they were dealing with only the most impactful class of mutations. The percent of effectively neutral mutations for their DFE is much smaller than Racimo & Schraiber, but it’s still enough to be a problem. I would say there is nothing at all unclear about this. The end result was still population collapse from genetic load.
Junk DNA cannot rescue life from GE because it doesn’t address the root cause of GE, which is the fundamental imbalance between deleterious and beneficial mutations.
It’s possible, but that seems a very odd paper to cite for that number.
Why cite a ‘consensus view’ from a paper on a different topic? Are there no experimental results available from papers on the same topic?
This looks like cherry-picking.
1 Like
To sum up from my perspective.
Kondrashov’s paradox is the disagreement between the models population genetics and the conclusions reached from observation. The models are simplified and are known not to catch the foul complexities of the matter so the rational response is to provisionally attribute it t0 a failing of the models - and indeed the relevant experts seem to agree on that and some are actively working on solutions.
The argument against this is to quote authorities who supposedly justify setting well=established facts aside in the face of theoretical difficulties. But the quotes provided certainly do not justify it and indeed the idea is so clearly irrational that one or two expert opinions would be insufficient anyway. Further, proposed solutions are dismissed out of hand with questionable - or even nonsensical - objections.
The matter is further confused by the issue of “genetic entropy”. Early on we were told that genetic entropy was Kondrashov’s paradox - but more recently we have been told that it is not. At the same time we weee told that fitness was only assumed to be a “rough proxy” for genetic entropy - and even then it is not a good one. From that It is clear that we do not have any empirical confirmation of genetic entropy since there is no measure for genetic entropy that would allow it, Indeed, it appears to me that genetic entropy is an interpretation of Kondrashov’s paradox based on assuming recent creation.
In short, the arguments against evolution and an old Earth presented here are bad apologetics which can be and should be dismissed.
7 Likes
The numbers account for transition bias, yes. Because there are still so many more ways of a codon mutating to encode a different amino acid, even with the bias nonsynonymous mutations are still more likely to occur.
Yeah that’s still not a DFE for beneficial nonsynonymous mutations. That number is still just a rough estimate for both silent, nonsynonymous, and intergenic mutations.
Further, the authors basically go on to state what I’ve been saying all along, that the DFE isn’t fixed, that as populations adapt the pool of available beneficial mutations diminish, and many more beneficial mutations are available to a poorly fit genotype:
On the other side, we found that among 48 random mutations, two were apparently beneficial. It is generally accepted that beneficial effects are ≈1,000-fold less common that neutral and deleterious ones (6, 39, 43). Therefore, it is striking that two of 48 random mutations were beneficial. However, this result is not so surprising if we recall that we used a chimera genome as template for our mutagenesis experiments. The template cDNA was assembled from clones of each of the VSV genes and intergenic sequences from two different sources. Whereas the N, P, M, and L genes were obtained from the San Juan strain of the Indiana serotype, the G gene was obtained from the Orsay strain of the same serotype (31). At the amino acid level, the divergence between the San Juan and the Orsay G proteins is ≈5%. The question is whether this difference precludes an efficient interaction between the Orsay G protein and the rest of the gene products from the San Juan strain. This being the case, many different possible ways to optimize such genomes are available. Furthermore, the ratio of beneficial to deleterious mutations depends on the degree of adaptation of the virus to the laboratory conditions, which in this case is minimal.
Curiously that paper also cites papers demonstrating the reality that as background fitness goes down, mutations become less deleterious and more beneficial. A phenomenon you have consistently failed to comprehend how affects your conclusions with your fixed DFEs.
You seem to think diminshing returns epistasis means that a beneficial mutation has a fixed maximum magnitude of effect, and only gets smaller in the context of another beneficial mutation (slowing adaptation), but recurrently fail to see that it’s effect also gets larger when background fitness gets lower.
That means, in the simulation you did, when your population’s mean fitness declines, both the frequency and mean effect-size of beneficial mutations should increase. This would have the effect of slowing and, depending on other factors like population size of course, possibly halting or even reversing any decline altogether (after all, populations demonstrably do adapt).
You’ve got a snapshot of a DFE at a particular fitness level with a large junk genome (explaining why so many mutations have such small effects), mistakenly thinking this should imply a decline because your simulation does not take the rising frequency and magnitude of effect of beneficials into account when fitness gets lower.
We’ve finally reached a point of agreement.
2 Likes
You seem to believe a 10% functional genome (representing no more than 300 millions bp) is enough to take on the informational challenge associated with building and maintaining a human body. I think that as a default position, it is unreasonable. As to what those criticisms might be, according to @John_Harshman at 221, it seems that has to do with the fact that junk DNA isn’t conserved within or between species. Well, this argument for junk DNA doesn’t seem persuasive to me. The conservation-based definition of function was largely built on protein-coding genes. Applying it unchanged to a regulatory meta-information layer — which by its nature may operate through more flexible sequence rules — begs the question of whether the same criteria apply.
Well, I am spamming with the notion that it is unreasonable to think that 90% of the human genome is composed of junk DNA. On this, I share Cech’s view. Would you say that Cech is spamming crap?
No, not really. Cech’s point is mainly a historical one, noting that in the past, the function of other class of RNAs such as telomerase RNA, microRNAs and catalytic RNAs was unknown and so could also have been dismissed as noise or junk, the lesson being to never underestimate the power of RNA.
For your claim to be true, it would be necessary to assess the concentration of these lncRNAs in all cell types longitudinally during development, childhood and adulthood and across different environments. Do you think such requirements were met ?
So your view seems to be that these lncRNAs correspond to non functional transcriptional noise. But what do you do with the observations reported in the article below:
https://academic.oup.com/gbe/article/18/5/evag042/8498184?login=false
From Sanjuan Moya Elena The distribution of fitness effects caused by single-nucleotide substitutions in an RNA virus
It is generally accepted that beneficial effects are ≈1,000-fold less common that neutral and deleterious ones
It should be noted that that is not the ratio of beneficial to deleterious. Given ~ 10% or so of DNA involved with regulation or coding, that would work out to 100:1 without neutral, which seems be in line with other studies. Of course, the propagation of both in a population is then subject to negative and positive selection.
2 Likes
Well, since you seem to lack the vast majority of relevant information…
I’d say that’s just wrong.
I’d say that resorting to hearsay instead of Cech himself is crap.
1 Like