Correcting a Quote Mine on Deleterious Mutations

Thank you. But sometimes I do think I understand much more than I do. If I really overstep what I understand, you guys will let me know. :sweat_smile: I’m trying to do better so it doesn’t happen as often.

Bacteria are a different ballgame.

Do they?

1 Like

Some nearly neutrals must be beneficial. Saying otherwise is a rejection of basic mathematics.

No. And I quoted specifically what I did intentionally, as we can’t demonstrate anything about them conclusively. Effect size of nearly neutrals are far too small to be conclusively established at all. We can infer the exist from certain biochemical principles (excessively large genomes are slightly deleterious because of increased metabolic costs of reproduction, say), but these are below the capacity of reasonable effort to detect.

You can just read what they actually say… Reduced selection, to the extent they say it at all. Implying of course that a) normal levels of selection would not result in fitness reductions and b) none of it applies to any other species necessarily, and c) the reductions will reach a new equilibrium and then stop.

I guarantee this causes you less harm than your tendency to think people like Paul or Jeanson know much more than they actually do.

And yet it does apply to viruses, in spite of even higher population sizes… Odd, that. All most like they’re just making stuff up to fit whatever they need as they say it.


Not with respect to the DFE of mutations in functional or junk DNA, zero evidence that they deviate*. The only main difference between bacteria and large multicellular eukaryotes that could meaningfully impact the DFE of mutations measured across their total genome would be that most large multicellular eukaryotes have a lot more junk-DNA.

But of course at the physical and biochemical level, bacteria still operate on the same physical principles as larger and more complex eukaryotic organisms do. They still have enzymes, they still regulate their expression with DNA binding proteins that initiate or inhibit transcription, they still have phospholipid bilayer membranes that provide barriers between the inside and outside of cells, they still employ proteins to transport waste products out and nutrients in. They still need to eat to survive, and grow and divide to reproduce.

So for all these same reasons, bacterial protein coding genes and promoter and enhancer regions would have local and global optima that would entail essentially similar DFE of mutations, which would also depend strongly on how adapted they already are and change as a function of overall fitness.

Oh and as I have said numerous times, since neither Sanford, Carter, or Price can give me an equation for the rate of fitness decline due to Genetic Entropy that has a factor called “simplicity”, their excuse that “simple” organims should suffer less GE is totally meaningless and vacuous. It gets even worse when, as @CrisprCAS9 points out, by any intuitive account viruses are much simpler than bacteria, and yet these are supposedly succumbing due to GE.

*Edit: Forgot to mention, bacteria generally have huge population sizes, which would imply that mutations of smaller effect are visible to selection.


Bacteria are a different ballgame.

That is the claim–yet if you look closely, the citations often used to support GE are derived from experiments with bacteria. Paul has regularly quoted bacterial accumulation assays as evidence on Reddit, not sure if he does here.

The reason why bacteria is a “different ballgame” is because GE doesn’t make the correct mutational predictions in the experimental system i.e.–growing bacteria in the lab.

GE is premised on genetic drift allowing deleterious mutations to accumulate in an organism’s genome which then cannot be prune by natural selection. This is the basis of Kimura’s and Ohta’s work. They recognized that non-neutral mutations could propagate in a population under special circumstances that allow genetic drift to overpower natural selection.

This occurs when the selection coefficient, s, is much lower than 1 divided by the effective population size, Ne:

|s| << 1/Ne

For bacteria, E. coli for example, Ne = 1,000,000,000 or 1 billion. So, a deleterious mutation cannot act neutral (and therefore will be selected against) if it decreases the fitness by even 0.000000001.

We can see this experimentally too, which is why GE proponents have to make up special rules for bacteria. It additionally happens with viruses because their Ne is very large–which is why the H1N1 narratives as “evidence for GE” are ridiculous (among a litany of other errors).

In the case of humans, Ne is between 10,000 - 30,000 meaning that natural selection can remove anything with |s| << 1/2(Ne) [2 because diploid] or 0.00005 - 0.0000166.

Essentially, GE requires
a) that an overwhelming majority of mutations are deleterious [not supported by data] and
b) that natural selection is never allowed to prune mutations [unsupported]



It’s not in his book, but this paper is referenced on Sanford’s website as a new development. It doesn’t come to that conclusion - it’s sort of opposite of everyone else and what you were saying - granted they say selection will increase. But all of this seems to be begging the question.

We estimate that humans and chimpanzees have accumulated approximately 140,000 slightly deleterious mutations each, mutations that would have been eliminated by selection in murids. These mutations have small effects, since it can be inferred that they have selection coefficients less than 1/ Ne for hominids, i.e., less than 10−4. It should be noted that it is unlikely that the mutation accumulation is due to a recent relaxation of natural selection in humans due to an improvement in our living conditions [32], since the time of this improvement is short relative the overall length of human evolution. We would not expect the decline in fitness to continue indefinitely, since the absolute strength of selection on new mutations, both advantageous and deleterious, may increase as fitness declines [33]. Furthermore, this accumulation of deleterious mutations may have been compensated in part by adaptive substitutions in gene expression control regions and elsewhere in the genome.

It seems like this is a silly argument when I can look up basic biology info about replication and find that selection obviously plays a role as they explain below.

Yes, but also this:

Bacteria have an amazing growth rate. The entire world population of a species like E. coli turns over very fast (perhaps once per hour). Trillions upon trillions of these cells die for many different reasons each and every hour. Thus, this may be a system where natural selection can actually halt the inevitable decay. Why? Because any mutation that confers even a small disadvantage (and most do) can be removed through differential reproduction, given enough time. (Time in this case is measured in generations.)

@Rumraket thanks for sharing this paper with me. The irony is that it has some of the same authors as the paper above, even though they seemed to refute some of their same research. But all the fitness distributions in the graphs were uneven.

What I thought was key:

This raises a crucial question: can we ever really know what the DFE is? For a simple organism like a virus, this does seem to be possible, because most mutations have large effects that can be assayed in the laboratory. For most other organisms, and particularly
for multicellular organisms, quite the opposite is the case; most mutations, even if they are deleterious, have such small effects that one cannot measure their fitness consequences. Furthermore, the environment in which most organisms live is probably sufficiently complex that laboratory assays give only the crudest measure of fitness. However, comparative methods using DNA sequence data potentially allow us to circumvent both of these problems. By examining the pattern of polymorphism in a species, we can estimate the effects of mutations with very small effects and we can infer the overall fitness consequences of mutations in the environment in which the species evolved. Furthermore, if we are prepared to sequence hundreds, if not thousands, of alleles then we can, in principle, measure the DFE for both strongly and weakly selected mutations. So, the DFE is knowable, but uncovering it might require a large amount of sequencing and effort.

Isn’t this what Sanford and Carter did?

Isn’t this what they’re doing with SARS-CoV-2? No undisputed increase in fitness in any mutant so far?

So…this is the reason why I think GE has merit based on what I’ve seen in the forum so far:

  • When human degeneration is pointed out as evidence of GE, then scientists say natural selection must have acted (this is begging the question based on your/their view of evolutionary theory)

  • I haven’t seen proof of how scientists think natural selection acts or how scientists have researched/posited it, other than how Sanford says they do.

  • Sanford convincingly (IMO) refutes how scientists think or have posited how natural selection acts.

  • I have seen no one try to state or prove he’s wrong on natural selection.

  • Others bring up beneficials and their distribution, and we go back to Bullet #1

For those that haven’t read the book, this is the point I’m up to and Sanford goes beyond the Kimura’s no selection zone.

As always, let me know what I’m getting wrong or missing. This is my non-scientist brain, trying to understand a science world. It’s interesting :slight_smile:

Wow, Sanford is referencing a paper showing the expected result of a century of population genetics? Stop the press! Hominids have lower effective population sizes than mice and as such must take on an increased mutational load, as they say in the very first sentence you quote. That’s it. That’s not evidence that Hominid genomes are on an indefinite fitness decline, that’s evidence that they have lower Ne.

Hey @thoughtful, please explain to me how growth rate, that is the number of new individuals produced by unit time, makes a difference to how many deleterious mutations there should be, or whether they are visible to selection. Pro tip: It does not.

Yes and that is not a controversial result. In total, more mutations are neutral and deleterious than are beneficial. What is not supported by any research is that the distribution is skewered and fixed such that deleterious mutations of very small effect will accumulate indefinitely.

That last sentence is the key point of contention that really matters to whether GE is true or not, and there is no evidence that it is true. Every word in the sentence matters. I’ll repeat it:
What is not supported by any research is that the distribution is skewered and fixed such that deleterious mutations of very small effect will accumulate indefinitely.

Sanford, Carter, and Price can cite no study that substantiates the claim that the distribution of fitness effects of mutations is skewered and fixed such that deleterious mutations of very small effect will accumulate indefinitely. This result only and exclusively obtains in physically unrealistic mathematical abstractions. But that is all they are, abstractions.

That’s what the tried to do with the H1N1 paper, but failed at doing. @dsterncardinale has this nice video that explains why:

There’s that word “undisputed”. If someone, anyone, disputes it, that gives you some excuse to dispute it too? Can that someone be wrong?

You mean acted in the past in ways it cannot currently act due to advances in medical science, agriculture, and technology? Well they’re right to say that, and that’s not begging the question. It’s flat out ludicrous to suggest otherwise. Simply go compare the infant mortality rates in afghanistan to those in, say, Belgium. Or look up historical rates of death in pre-industrial times due to famin and disease, and then compare them to now.

It’s not clear what the question you’re asking even is. What is it scientists are supposed to “prove” about how they think natural selection acts?

When he says what, specifically, about natural selection?

And bullet #1 was the factually unassailable reality that selection is relaxed in modern industrialized nations, so how does this confirm anything wrt GE?

Well that’s certainly a pretty figure Sanford has drawn there. Why are there no actual numbers on the axes? What experiments did he do to derive the size and distribution of both the curve and the “no selection zone”? Wait, none? Hmmm.


That’s the defence used by quote-miners.

Having just read the paper (here) it’s clear from the title ("…mutations affecting fitness") and the other sections which talk about protein-coding genes and amino-acid changes, plus the methods used in the studies they reference, that they are talking about mutations that have observable effects, not about all mutations including nearly-neutral ones.

So yes, they said what they said. But what they meant is changed by the context you have omitted.

That’s quote-mining.


I’m sure this has been pointed out earlier, but in case anyone missed it or needs a reminder, here is GE refuted in a single graph courtesy of the Lenski LTEE:


You can assert that all day, it’s still not true. Nobody has demonstrated any context from that paper which invalidates the clear meaning of the statement I quoted. I have read the whole paper myself. You can read the extra quotes provided in the OP in an attempt to say it was a quotemine; they fall far short of demonstrating any such thing.

Perhaps not just reduced selection but also changed selection. For example, there’s decent evidence that in Western Europe, at least, there is currently selection against alleles associated with higher intelligence (e.g., if one wants to consider that degeneration.


Correction: You have not yet admitted that this has been demonstrated.

The truth of a statement is not determined by whether you accept it.


From the abstract:

t therefore seems likely that the lack of conservation and increased rate of gene expression divergence are caused by a reduction in the effectiveness of natural selection against deleterious mutations because of the low effective population sizes of hominids.

So they say it is a reduction in selection in hominids. From the conclusions:

We estimate that humans and chimpanzees have accumulated approximately 140,000 slightly deleterious mutations each, mutations that would have been eliminated by selection in murids.

So it wouldn’t be happening under normal levels of selection (my point ‘a’) and it isn’t happening in murids (my point ‘b’). From the bit you’ve quoted:

We would not expect the decline in fitness to continue indefinitely, since the absolute strength of selection on new mutations, both advantageous and deleterious, may increase as fitness declines

So the reductions will reach an equilibrium and stop (my point ‘c’).

They in fact say exactly what I said population geneticists say, point for point!

Thus, this may be a system where natural selection can actually halt the inevitable decay. Why? Because any mutation that confers even a small disadvantage (and most do) can be removed through differential reproduction, given enough time.

Viruses have population sizes several orders of magnitude larger than bacteria, so if bacteria are immune to GE then so too must be viruses.

because most mutations have large effects that can be assayed in the laboratory.

Meaning that GE can’t possibly apply to viruses, as all would be under effective selection.


Please stay on topic, of this quote mine. Do not introduce different and irrelevant information. It makes so much work for the @moderators to split this, and if we don’t it creates a mess for readers.

1 Like

Help me understand why what you’re saying is not contradictory with what @CrisprCAS9 says.

I see you saying we do know DFE (more mutations are neutral and deleterious) and natural selection is relaxed in humans which explains why GE is not true.

I see him and others saying we can’t know DFE (therefore GE is wrong and you’re wrong) and natural selection also wont see those same mutations (so they will accumulate). But he also agrees natural selection is relaxed somehow and would normally see those mutations. To me, it seems like you contradict each other and sometimes yourselves.

I’ll try to watch it later, but it was already obvious from the thread he didn’t understand their argument so I’m not sure why I would be convinced of his critique of their paper.

It’s mildly amusing that we’re talking about an increase in fitness for 1 mutant out of 1000s and you see the need to argue that it matters whether it’s disputed or not. To me, this makes scientists look unscientific.

Is there any research to show the extent to which it is relaxed? Or what would be expected if it was not?

The context of the study was very clearly laid out and the authors are very clear in their position elsewhere. Neither author believes that all mutations are overwhelming deleterious. Both authors correctly believe that amino-acid altering mutations are largely deleterious–this is not the contextual claim presented by the quote.


Then give a quote that shows this, don’t merely assert it. Based on what I’ve read from them, they would seem to be in agreement with what Kimura wrote. They acknowledge that most mutations are deleterious, but they are holding out that rare beneficials may somehow overtake the decline.


It also bears mentioning that you’re stepping outside of what it means to “quotemine” something.

[Emphasis added]

It’s only a quotemine if the authors are clearly saying something different in the context of the actual quote! Not “elsewhere” (in other writings).

If what they say here is in conflict with what they say elsewhere, then they are either guilty of contradicting themselves or of being misleading (whether intentionally or not is a different issue).

Paul, I did. It is in this exact thread two times with clear explanations. I presented this to you 10 months ago when you made this exact claim. In fact, my initial response was copy and pasted from our discussion. Here’s what Eyre-Walker and Keightley have to say:

The DFE differs between coding and non-coding DNA.

These observations might also indicate that most amino-acid-changing mutations are deleterious; for example, if we infer that at most 30% of non-synonymous mutations are neutral in humans, this implies that at least 70% are deleterious.

In mammals, the proportion of the genome that is subject to natural selection is much lower, around 5% (Refs [55]–[57]). It therefore seems likely that as much as 95% and as little as 50% of mutations in non-coding DNA are effectively neutral; therefore, correspondingly, as little as 5% and as much as 50% of mutations are deleterious.

Eyre-Walker, A., Keightley, P. The distribution of fitness effects of new mutations. Nat Rev Genet 8, 610–618 (2007).

Based on what I’ve read from them, they would seem to be in agreement with what Kimura wrote.

I’m more than happy to talk about Kimura too. I’m would be giddy to link your deleted post from Reddit when you realized I was correct and posted under an alternative account in /r/genetics. We can discuss Kimura’s 1991 paper again.

Absolutely not. You attempted to convey a consensus from the authors that most mutations are deleterious. You did this by ignoring the amino-acid alterations under consideration for the study quoted. I demonstrated this is not the view held by the authors by citing their review article on DFE. Would you like me to find quotes from you that mischaracterize your views?

Do Eyre-Walker and Keightley believe that most mutations are deleterious?


Can you send them an email with a focused question, that they could give a public answer to?

The key to your question is the first sentence of mine you quote.
It begins with in total [more mutations are neutral and deleterious] [than are beneficial].”

I am not making a claim about the relative proportions or magnitudes of the [neutral and deleterious] ones, merely that they [in total] outnumber beneficials.
Just hypothetically, they might outnumber beneficials 80d+n:20b, or 90d+n:10n or even further, depending on how far into high levels of fitness the population already is.
But then the [neutral+deleterious] fraction might itself be split [70d:30n], or [50d:50n], or [30d:70n] or w/e.
I fully acknowledge that I don’t know the distribution within the neutral fraction, but even if it really is somehow close to how Sanford imagines, I’ve already given very good arguments (supported by real empirical evidence) for why his conclusion of inevitable and indefinite fitness decline physically cannot occur.

Uhm, no, just no. Just flat out false. Watch it.

It makes scientists look unscientific that they really care about being precise and correct in their characterizations of real data? Then perhaps what you find mildly amusing is itself an argument that you’re not being all that reasonable.

Hey, I see you also skipped right over me asking:

I invite you again to consider taking a look at both extant and historical comparisons in measures of mortality rates for human beings, between industrialized and non-industrialized nations, and pre-industrialized and current time periods. For example, take a look at table 2 in this paper:

Compare those historical values to the modern rate at the bottom. From about a quarter to roughly half of all children died before they reached adulthood. That’s an enormous selective pressure.

Yes, there is a vast literature on this, and among it are also publications that put these numbers into an evolutionary context. Population geneticists such as Michael Lynch have pointed these things out: