Side comments: Genetic Entropy will be debated once again

[UncensoredPilgrims (Paul P)]

@talkpopgen

When I said “Paul rejects such a thing exists,” I meant simply that you don’t accept humans have been around long enough for such an equilibrium to be reached.

That’s not really a relevant point. The relevant point is, “what figure should be used to calculate the percent of the DFE which is effectively neutral”? That is the point relevant to Kondrashov’s pardox. The correct figure would be long-term, not short term. The long-term Ne is on the order of around 10,000, and certainly not in the billions whatsoever.

@talkpopgen

Absolutely. But genetic entropy makes two predictions: 1) we should be extinct already and 2) we are heading towards extinction. For (1), long-term N_e is what we should be concerned with; for (2), I argued it was variance N_e, as this determines the rate of genetic drift happening in the contemporary population. Partitioning these two predictions is not a red herring; I’m merely addressing these one at a time.

That’s fine, and I am perfectly fine saying that the present-day large population sizes of humans will have the effect of slowing down fitness decline compared to if the industrial revolution never happened and we stayed in much smaller, more isolated groups. But again, it’s number 1 that is really relevant to the question: “Are mutational effects a problem for evolution?”

@talkpopgen

I’m very happy to update to an estimate nearer Felsenstein’s, though to be precise we’d need to consult global life-tables.

Thank you for the honest concession there.

@talkpopgen

I never claimed that. I said the exact opposite.

No, you did not say the opposite. You made the claim in one of your videos addressing GE that we should be looking at recent Ne rather than long-term Ne. And for all the reasons I have stated repeatedly, that is simply not the case.

@talkpopgen

I have the same degree and have published in the same journals as both these people. Except somehow my credentials matter less.

This isn’t a battle of credentials in the first place. Yours certainly don’t matter any less, but if you disagree with them then that’s what you should say. You shouldn’t tacitly allow people to assume that you are agreeing with what they have said, but then falsely accuse me of quotemining.

@talkpopgen

Empirically, Couce et al. (2024) showed the DFE evolves as the population does, with mutations that increase fitness becoming rarer as a population adapts to a new environment (in this case, laboratory test tubes).

I don’t question that as a real phenomenon. Diminishing returns epistasis is very real. It is the claim that the environment can change, and suddenly the ratio of deleterious-to-beneficial can then swing wildly and allow for great leaps of fitness (and corresponding great leaps of functional complexity, such as is needed to turn bacteria into people) – THAT claim is the one which stands as a just-so story without a shred of evidence to support it.

@talkpopgen

This is the exact point I was making in the debate regarding traits with an omnigenic architecture, which seems far closer to biological reality than assuming selection acts on individual mutations.

You are promoting a theory that essentially throws out the theoretical basis of modern population genetics, based on neutral theory, and asserts that NS is somehow able to “see” even the tiniest of mutations and act on them. But that is not what Kondrashov or Lynch believe, and it is not remotely what you attempted to model using SLiM.

By conceding that selection cannot in fact act on individual mutations, you are conceding that you have no mechanism to purify out the vast majority of them, which are demonstrably damaging.

This claim is not an overturning of Kondrashov’s paradox in the least. Instead, it represents a handwaving away of the problem entirely.

Another way to put it is this: you’re putting the cart before the horse, and appealing to phenotypic differences, while ignoring that mutations have to accumulate slowly, generation after generation, and most of them have no discernable phenotypic effect at all.

Kondrashov’s paradox does not depend on any comparison to a hypothetical individual with no mutations. Certainly, Kondrashov himself never suggested he believed in such a thing. It’s simply the logical result of an understanding of the DFE, and the end result of what will happen when you allow slightly deleterious mutations to accumulate over time with no means of purging them. That is what I modeled using SLiM.

I’m going to let you have the last word on here if you want to take it, since it is not my desire to sign up for an indefinite back-and-forth on the matter in the forums.