GE of the coronavirus doesn't make sense

I was thinking about this (and other) threads we had not too long ago:

And I got the impression that some of the Genetic Entropy proponents around here might be suspecting that Genetic Entropy constitutes a good explanation for India’s purported “plummeting COVID-19 cases”, or declines in the rates of deaths and hospitalizations reported in different countries all over the world.

Before going on, let’s remind ourselves that GE is the idea that there is relatively constant, and vast(the overwhelming majority) class of mutations with deleterious fitness effects so small they are effectively invisible to natural selection, and that these will pretty much always predominate on net effect such that, through mutation accumulation the mean fitness of all carriers in the population will inevitably slowly decline, until mean reproductive rate drops below 1.0, in which case the population size will decline and the species eventually goes extinct.

Okay. Now let’s look at the Corona virus. The most genetically divergent strains are something on the order of about 50 mutations different from each other, and in the area of 15-30 mutations different from the first identified variants making the rounds back in december 2019 or thereabouts. A few tens of mutations in total are separating the different strains.

That means, if you think Genetic entropy is causing the declining Covid-19 cases and hospitalizations, that a Coronavirus carrying approximately 30-50 mutations of effectively invisibly small fitness effects, is decreasing it’s fitness from whatever it’s ancestral value was back in end 2019, to values below 1.0.

Now obviously that does not make sense. How can ~50 mutations of invisibly small fitness effects (s ≈ -0.00001 or less?) add up to a fitness decline so significant that the virus is supposedly going extinct?

It seems to me the most obvious explanation is that more widespread implementation of precautionary measures is what is taking effect. There’s just no sense saying the virus is somehow succumbing to a fitness decline due to mutation accumulation, when (first of all), the variants now known to dominate in most places are radically more infectious than when the pandemic began, and when we’re only dealing with such a small number of total mutations. It simply doesn’t add up to think of this as the result of GE.

1 Like

I don’t remember that. For my part, I suggested they may have reached an immunity level where it’s circulating slowly until they reach herd immunity. That may be the case in Florida as well, along with the kind of climate right now the virus doesn’t do well in - we may see a slight summer peak around the country again when everyone is inside with air conditioning.

I don’t think we’ll see the effects of GE for several years. I believe Sanford and Carter were saying the effects weren’t significant for several years, in fact there was a higher second peak with the Spanish Flu, IIRC? Also, coronaviruses mutate more slowly than the flu?

I didn’t feel like commenting on it when I made the thread linked below, but I was sort of curious if anyone would pick up on it… When I read more molnupiravir after seeing the headline, it seemed like the kind of antiviral that Sanford and Carter were advocating would be helpful especially in a pandemic. If it continues to do well in trials, that might be better evidence of GE than anything we’ll see in the immediate future. But that’s only if I understood its mechanism of action correctly. :relaxed: If not, then I’m sure someone will correct me.

According to the analysis of S&C, the accumulation of point mutations within the H1N1 virus occurred at 14,4 new mutations / year, which is close to what we saw this year for the coronavirus.

The number of accumulated mutations in one year is probably to small to have a clear visible effect on fitness. But if the coronavirus go on to accumulate ~ 20 mutations / year, I would bet that in 10 years, after having accumulated 200 mutations, a reduction in fitness will be seen.

Okay, thanks for clarifying that you weren’t thinking along those lines. At least my post here above does not apply to you then.

Trying to put on a hat here and see this from the perspective of someone who believes in GE, I believe that makes much more sense than to say (given what I’ve stated above) that the virus is currently succumbing at any measurable level to GE.

This does raise a question about when exactly a GE proponent would expect it to take measurable effect, and in turn this has implications for whether GE is actually a practically testable scientific hypothesis.

I didn’t feel like commenting on it when I made the thread linked below, but I was sort of curious if anyone would pick up on it… When I read more molnupiravir after seeing the headline, it seemed like the kind of antiviral that Sanford and Carter were advocating would be helpful especially in a pandemic. If it continues to do well in trials, that might be better evidence of GE than anything we’ll see in the immediate future. But that’s only if I understood its mechanism of action correctly. :relaxed: If not, then I’m sure someone will correct me.

Again here I think we need some way to distinguish the putative predictions of GE from those of canonical evolutionary theory. I think there are many population geneticists who would agree that there is definitely such a thing as a mutation rate that is too high, and that it is possible such a rate of mutation can be instigated with drugs. It is not enough to merely say that there is some circumstance under which mutation accumulation will cause fitness decline because mutation rates can be made artificially high with drugs.

GE is the claim that natural evolution is impossible because fitness decline until extinction is the inevitable fate of all natural populations under their normal ranges of rates of mutation, as natural selection is powerless to prevent the accumulation of mutations of invisibly small deleterious effects.


Sanford and Carter conflated mutagenic drugs as applied to individual case management, to somehow being a tool to bring about accelerate the attenuation of virulence at the population level. The course of viruses replicating in ones body, and what circulates in a population, are dramatically different scenarios.

Such drugs have long been investigated, well before the pretentious posturing of Sanford and Carter, who have contributed no insight or anything of value to lessen misery or save lives. Of the over ten thousand preprints available concerning COVID now available, tell me how many pay any credence to Genetic Entropy?

We can identify the exact gain of function mutations in coronavirus which appear to be correlated with increased morbidity, contagion, and escape. Why is it we never get from the likes of Sanford and Carter any identification or list of specific deleterious mutations, for instance stipulating what substitution at what position is deleterious because it results in less host affinity or other function?

Does Sanford really think that epidemiology is some sort of passing fancy that you can just waltz into and tell thousands of people who have dedicated their lives to investigating viral disease that they have missed the big picture? He is a plant biologist who has published less than a handful of papers relevant to epidemiology, which he himself states was really about Genetic Entropy, and having next to no influence or citation in the field. If Sanford had any competence or serious interest in making a contribution, he would not be off writing books on paleoanthropology and promoting drivel about the math of lifespan decay on the creationist speaking circuit.


If GE is correct, there’s no need for herd immunity.

I am positive that we will never see any effects of GE, because there is no evidence that it exists.

I believe that S&C were fudging the definitions of strains and subtypes throughout.

So do we, but they claim that we are undergoing GE, so that’s irrelevant.

Yet, they, being part of an organization with a $2M/year budget, aren’t lifting a finger to do any science. Why is that?

I don’t see how it could ever be evidence of GE.

H1N1 is a subtype. There is no entity called “the H1N1 virus.” That is the main technique they are using to fool you (and maybe themselves).

Several have already had a clearly visible effect. They increase fitness, which is why they are displacing virus lineages lacking them. This is all neutral and Darwinian evolution happening right before our eyes.

Indeed. It clearly isn’t a scientific hypothesis.

It certainly has nothing to do with the hypothesis of GE.

And why aren’t there any authored by S&C? What could be a more dramatic demonstration?

I would edit that to:

If Sanford had any real belief in his competence or serious interest in making a contribution, he would not be off writing books on paleoanthropology and promoting drivel about the math of lifespan decay on the creationist speaking circuit.


It is good that you included that qualifying statement, as there is very big and unjustified assumption at play.

That an antiviral drug’s efficacy is due to inducing mutation does not validate GE. The thesis of GE is that the relentless accumulation of slightly deleterious mutations over multiple generations results in ultimate collapse. The object of mutagenic drugs is to riddle the replicating viruses with as much lethal damage as quickly as is possible without killing the patient. The intention is to reduce the viral load by highly selectable mutation. If the mutations were not selectable, by the time GE caught up the patient would likely be long gone.

The objection to GE is not based on the idea that mutational collapse cannot happen. It can due to selectable mutation, and as such inbreeding is a concern in conservation in small populations. But as @Rumraket has explained repeated, there is no justification for Sanford’s distribution of fitness effects of mutations. Sanford downplays selection on the deleterious end, does not consider that a significant proportion of mutation is likely completely neutral, and dismisses positive mutation.

1 Like

The only thing I have to say about the idea of genetic entropy and viruses, after all the ink spilled and hours spent on the topic, is this: “haha, no.”

The key thing from GE is that deviation from some “optimal” genome must on net be negative. Which means results like this should be impossible.

And don’t get me started on that H1N1 paper. El Oh El.


The problem is that this logic applies not just to SARS-CoV-2 in humans, but the ancestral virus in bats. It didn’t poof into existence in the fall of 2019. That same viral linage had been replicating and mutating for millennia before then.

And then there are human viruses that have been around for longer. HIV. Polio. No GE there. Weird.

And organisms with larger genomes, large population sizes, and short generation times, be they bacteria or mice.

The hard part with GE isn’t picking out one case where it should apply in the future; it’s explaining away all the cases were it doesn’t in the present.


Would you and @thoughtful agree that stated even more simply, error catastrophe is a real, documented phenomenon, but GE is the hypothesis that error catastrophe is inevitable for any population?

Those look like fair representations to me.

I wasn’t the subject of the question, but fwiw, I would dispute this. Mathematically, it should exist. It’s easy to model it. But in practice…I’m not convinced it’s ever actually been demonstrated.

And before anyone runs off to link Crotty et al. 2001, lemme tell you about all the other stuff ribavirin does that weren’t controlled for. I wrote up a long-ish thing on this question not too long ago. I agree with Summers and Litwin (2006) when they say

While a detailed critique of the literature in this field is beyond the scope of this commentary, we find that, in general, experimental support for error catastrophe is marred by the failure to propose or test alternative explanations for the results and by inadequate precision in the data.


Fair enough!

Graur used that model to estimate the functional fraction of the human genome:

He’s since issued a correction to that paper based on this work, on which he was a reviewer.


Ok, have you read about molnupiravir and do you think this also true for it?

I would agree I think, but so far I’ve only seen the first clause agreed to by a few scientists - those that say mutations should be embraced by using these sorts of anti-virals. At one time I quoted an article that said something similar to error catastrophe can almost happen.

I didn’t notice that mechanism in anything I read about this drug. Feel free to show it is the case.

Explanations exist for how RNA viruses may have resisted GE.

My question was general, not specific.

Sopecifically, I fail to see why antivirals that target the viral RNA polymerase are in any way relevant to GE. We already know that increasing chain termination or increasing substitution mutations attenuates viral pathology, because RNA viruses, including both influenza and coronaviruses, do this naturally.

It’s a feature, not a bug; understanding this is sufficient to recognize the inherent absurdity of GE. It’s telling that those promoting GE ignore this.

The relevant entities are called defective interfering (DI) particles. This is so old that there is a Wikipedia page explaining them:

and far older because I did a first-year lab rotation in 1979 doing RNA fingerprinting (this was before routine sequencing) of DI particles generated by infectious bronchitis virus of chickens, the first coronavirus ever isolated.

As for “embracing” mutations, DI particles have long been considered and tested as clinical interventions with little success AFAIK.

1 Like
  1. Influenza’s mutation rate doesn’t differ between hosts.

  2. In order for that explanation to work, they’d have to be dormant except for human outbreaks.

I mean, we can do the math on this. Sanford’s central claim is that deviation from some optimal state is, on net, negative. We have the mutation and substitution rates, and the genome sizes, along with reasonable approximations of the population sizes. So its pretty simple to calculate how long it would take for the population to sample every possible mutation, at which point no amount of selection, recombination, or (for influenza) reassortment can save it.

This has been done experimentally as well (eg Springman et al., 2010 “Evolution at a High Imposed Mutation Rate”), and across the board what we see contradicts Sanford’s claims.

Also, bacteria, mice, etc…they hanging out dormant as well? What’ the excuse there?


It’s almost as if this topic has never been discussed, and no one has ever given these issues a thought…

Ditto on this.

Sort of laughing, because I’m sure that you guys (meaning those who are not creationists) don’t like repeating the same explanations… obviously neither do I… But what’s funniest to me is we’re literally the same forum users who have had these conversations before and probably within the last 6 months. So it’s laugh or cry for me today.

Sure haven’t learned my lesson on not getting involved in GE conversations. :sweat_smile: But if anyone wants to answer the other parts of my post regarding topics not discussed before I would be happy.