Or GE analysis is wrong.
2 Likes
If current population genetics theories are wrong, then GE analyses are based on incorrect assumptions, since they are (allegedly) assuming otherwise. Therefore, GE analyses tell us nothing about anything if current population genetics theories are in fact wrong. On the other hand, if current population genetics theories are (more or less) correct, then GE analyses say that deep time is wrong. And that we know for a fact far stronger than any population genetics model, is definitely incorrect.
So, GE analyses are either wrong in their conclusions, or GE analyses are wrong in their conclusions and in their base assumptions. Congratulations.
3 Likes
Correct
Then again, according to deep time, GE is wrong, and there’s a whole lot of support for the former. Nor are GE and population genetics the same thing. Pop gen is a huge field that accommodates a mass of theory, not all of which, perhaps crucially, is contained within that GE simulation. So GE can fail without collapsing pop gen.
I’m sticking with my point: deep time falsifies GE without the need to discover what’s wrong with the simulation and without invalidating the entire field of population genetics.
6 Likes
Agree.
Agree
They are not the same thing, but the former is a development of the latter.
I disagree. If deep time falsifies GE, something I acknowledge, then something may be invalid within the field of pop gen, given that GE is a development of pop gen. Of course, there is the possibility that Sanford et al made an error when they developed GE from pop gen. But in that case, it is important to pinpoint the error, isn’t it?
GE is a claim about the consequences of a particular pop gen model. But pop gen is a big field, and it encompasses a great many models, with a great many possible parameters one might consider. Think of this analogy: if Newtonian mechanics is unable to account for the precession of the orbit of Mercury, do we then declare that physics is invalid?
It would be nice to poinpoint the error(s), but it’s not necessary to do so in order to know that GE is just plain wrong.
5 Likes
You don’t need to be “smarter”, you just need to have a basic understanding of evolution.
Two geographically separate populations that at one time were members of the same species later may have diverged into two different species. Since the process is gradual, there is no particular point at which it is possible to say that the two populations have become two different species. [1]
If “there is no particular point at which it is possible to say that the two populations have become two different species”, there is no point at which a specific first individual member of the new species existed.
Evolution, and speciation, occurs at the level of populations, not individuals.
Analogously, we are also unable to point to the first speaker of French – as the divergence from Latin was likewise gradual.
That somebody would consider themselves qualified to criticise evolution, let alone claim expertise in a subsection of population genetics, without understanding this very basic principle, is problematical in the extreme.
This level of basic misunderstanding Paul, not “religous ferver[sic]”, is why Creationist apologetics has zero scientific credibility.
5 Likes
Of course not, but it looks like the cases where you disapprove, your disapproval is fueled only by the AI-use producing results inconvenient to your beliefs.
No it doesn’t. That’s a completely baseless assertion.
Yes it is, and your disapproval is simply not of probative value. First of all it isn’t clear that AI is used at all, nor is it clear that doing so would result in detection of data-patterns that don’t actually exist. In fact I can’t find anywhere in the paper itself that any of the computational methods used in the paper or as any part of their data analysis involved AI, LLM, machine-learning algorithms or anything of the sort is at work at all.
AI is revolutionizing protein structure prediction from sequence data as we speak, and that isn’t being driven by any nefarious desire to somehow “come up with new discoveries that happen to contradict what actual human sceintists have said for many decades.” Even if it did imply that, those human scientists could have been wrong for decades and the AI models simply be correct.
After all, you yourself happen to think all the human scientists who have been saying for decades that evolution is true, are all wrong.
You’re just revealing your completely ridiculous double standards here.
Ahh no you have misunderstood the methods of the paper (as have those pop-sci article writers). The results of the deep mutational scanning is that somewhere around 1.5% of amino acid substitutions that occur in novel environments are beneficial. This led researchers to try to estimate another number, which is what fraction of amino acid substitutions that go to fixation in prokaryotes(or other species with very large Ne, such as yeast, are adaptive? That’s where they calculate the 99% of substitutions being adaptive number, from the fact that at Ne ~107 (a totally reasonable estimate for a flask of bacterial or yeast culture), a beneficial mutation can be one hundred thousand times more likely to go to fixation than a deleterious one.
Congratulations, you have discovered natural selection on fitness landscapes, you’ve just renamed the idea because you don’t want to admit evolution is real and that natural selection is demonstrably effective.
Yes yes, I know creationists have invented a new name for an idea that totally looks like natural selection but it somehow mysteriously isn’t, that has no mechanism, and is just a silly attempt to deny evolution while accounting for the same data with another name.
Straightforwardly doesn’t follow. The rate of change matters.
IF.
If environments change too quickly. Can that happen? Sure. Cataclysms like asteroid impacts, or man-made climate change that result in large-scale extinctions because many organisms can’t adapt in time. But that would have to happen everywhere on Earth and be in effect all the time to render evolution impossible. There is no reason to think that was ever the case.
3 Likes
In what universe do alleles = nucleotides?
Why? This would only seem relevant if something is assumed to be true until demonstrated otherwise.
In the case of GE, it’s never been demonstrated to be valid in the first place.
4 Likes
It would also assume that GE is wrong at only one single point – which would strike me as a fraught assumption.
It is not uncommon, with creationist arguments, to have trouble keeping track of all the different levels that they are wrong at.
3 Likes
It should be important to those promoting GE. But it isn’t.
It isn’t important for anyone else.
1 Like
The idea that environment may play a part in fitness effects seems to really get Paul’s dander up, but many examples can be referenced of mutations in large eukaryotic organisms which can be beneficial or deleterious depending on environment, and so defy permanent intrinsic value.
This is evident in that the reproductive success of given organisms can dramatically shift due to small environmental triggers, which is chaotic behavior. A example in my region is the pine beetle infestation, now abating, that devastated the local lumber industry. A few milder winters shifted the locus and allowed an explosion of infestation. By the measure of reproduction, the same beetle that was always present, abruptly enjoyed an more optimum fit between its lifestyle and its environment.
Environment can vary fitness effects in humans as well. High altitude adaptations of hemoglobin which are beneficial in Tibet would not be of any assistance in Denmark. Nor is rapid climate change a requirement for environmental variation; altitude, temperature, and precipitation differences are a matter of walking up a mountain range. Most populations already exist and mate over areas with environmental gradients which would exert differential selective pressures.
3 Likes
Since World War II, how many species can we say with confidence have gone extinct due to climate change?
I don’t know. Google it.
2 Likes
Ah! The sweet smell of probability distributions on a Monday morning 
I get p ~= 6.1295557 \times 10^{-17}, but I cheated.
“16” rather than 6 might be a typo? Either way we are still in the same ballpark.
1 Like
No, 0.6 is approx 1/1.6 (add zeroes as needed)
1 Like
Given a relative lull in this topic, I took the opportunity to query the Claude LLM to obtain more overview/context to the current state of play in population genetics theory and modeling. What I received was a lengthy laundry list of issues, as well as citations to a few useful sources. One of these was:
Which made the following comments that I found relevant:
The status of population genetics in contemporary biology is an interesting issue. Despite its centrality to evolutionary theory, and its historical importance, population genetics is not without its critics. Some argue that population geneticists have devoted too much energy to developing theoretical models, often with great mathematical ingenuity, and too little to actually testing the models against empirical data (Wade 2005). Others argue that population-genetic models are usually too idealized to shed any real light on the evolutionary process, and are limited in what they can teach us about phenotypic evolution (Pigliucci 2008). Still others have argued that, historically, population genetics has had a relatively minor impact on the actual practice of most evolutionary biologists, despite the lip-service often paid to it (Lewontin 1980).
[My emphasis]
&
This leads us to another facet of population genetics that has attracted philosophers’ attention: the way in which abstract models, that involve simplifying assumptions known to be false, can illuminate actual empirical phenomena. Idealized models of this sort play a central role in many sciences, including physics, economics and biology, and raise interesting methodological issues. In particular, there is often a trade-off between realism and tractability; the more realistic a model the more complicated it becomes, which typically limits its usefulness and its range of applicability. This general problem and others like it have been extensively discussed in the philosophical literature on modelling (e.g. Godfrey-Smith 2006, Weisberg 2006, Frigg and Hartmann 2006), and are related to population genetics by Plutynski (2006).
It is clear that population genetics models rely on assumptions known to be false, and are subject to the realism / tractability trade-off. The simplest population-genetic models assume random mating, non-overlapping generations, infinite population size, perfect Mendelian segregation, frequency-independent genotype fitnesses, and the absence of stochastic effects; it is very unlikely (and in the case of the infinite population assumption, impossible) that any of these assumptions hold true of any actual biological population. More realistic models, that relax one of more of the above assumptions, have been constructed, but they are invariably much harder to analyze. It is an interesting historical question whether these ‘standard’ population-genetic assumptions were originally made because they simplified the mathematics, or because they were believed to be a reasonable approximation to reality, or both. This question is taken up by Morrison (2004) in relation to Fisher’s early population-genetic work.
Claude concluded that, while pop gen modeling can produce “real, if approximate” results for narrower questions, larger questions like long-run absolute fitness dynamics remain intractable.
Based upon this, I’d probably take with a substantial grain of salt even narrow theoretical/modelling-based claims even from somebody with decades of expertise, and the very broad background in the subject needed to navigate what appears (from my layman’s perspective) a landscape littered with pitfalls.
The idea that I would accept very broad theoretical or modelling-based claims from individuals that appear to lack this length and breadth of expertise, a category that Sanford, Cater and Price very clearly fall into, when the contents of their claims seem counter-intuitive to myself, and are not accepted by mainstream pop gen, is ludicrous.
2 Likes
Pigliucci is a founding proponent of the Extended Evolutionary Synthesis, which proponents present as novel mechanisms augmenting the modern synthesis that serves as the foundation for population genetics. As a lay person ill equipped to adjudicate such a debate, for what it is worth the sense I get is that models based on genetic variation, selection, and drift, are well supported for a given locus, but nature in the wild is dynamic and chaotic, and models can therefore never fully capture or predict the future path. That a chaotic process based on the modern synthesis cannot be predicted, however, does not at all indicate different mechanisms are involved, only that over non-uniform natural conditions and over time, that variation, selection, and drift, will prove beyond the capabilities of a given baseline model. In my mind, the modern synthesis remains a comprehensive explanation, but acknowledgement of real world complexities is always in order.
Yes, I was more relying on the highlighted Wade 2005 sentence in that paragraph – I included the rest of the paragraph for context.
1 Like