Here are some wrong (yellow) and patently false (pink) claims from Svensson (2023)
Gomez, et al. address selective fixation of beneficial muts: this is NOT a model "in the absence of nat selection"
Nor is it a model requiring "unrealistically high adaptive mut rates": here we see effects over 10 orders of magnitude in mut rate
This is the latest installment of a 4-year disinformation campaign we are powerless to stop
I'll toot more about it tomorrow
@JoannaMasel #popgen #mutation
Continuing this thread on the latest in a 4-year disinfo campaign by
@EvolOdonata that we have been powerless to stop...
I explained the "blatantly false" parts, now moving on to just "wrong"
Must fitness effects align favorably with mut biases, as Svensson implies?
No, e.g., they don't align in Gomez or Yampolsky-Stoltzfus
Does the efficacy of arrival biases require drift, as Svensson suggests?
No, e.g., if the first beneficial variant to arrive is accepted deterministically, biases in arrival matter
Where does Svensson get this idea? He cites Lynch, but Lynch is using Bulmer's 1991 model where the worse of 2 codon types may fix by drift in small pops. The chance is inversely related to N_e . . .
That good-v-bad scenario is irrelevant to something like Yampolsky-Stoltzfus where the issue is which of 2 adaptive changes happen
Lynch (2007) invites this mistake by making a far-too-general claim about conditions for mut to influence "the direction of evo" using only Bulmer's 2-state (good v bad) model
Again, that does not cover the case of multiple beneficial options subject to arrival bias
I called out these errors when the preprint came out. They were not corrected. Previous attempts to correct Svensson's misinfo were rebuffed at TREE, and now Evol Ecol has refused a concise and highly informative rebuttal
https://ecoevorxiv.org/repository/view/4705/
The silver lining is that Svensson is building a documentary record proving our case for novelty: his ability to publish obviously false claims for 4 years shows that reviewers and editors know nothing about the theory
To learn more about the theory of arrival biases, check out wikipedia:
https://en.wikipedia.org/wiki/Bias_in_the_introduction_of_variation
Why would evo change not reflect u?
The classic answer from Fisher and Haldane (https://www.molevol.org/opposing-pressures-argument/) is that bc u is small, mut pressure is a weak force unable to influence the course of evo, important only with abnormally high u and no opposing selection
This way of thinking led to the belief, still widespread today, that mut effects will be important only when selection is absent (neutrality)
That's one answer
But F & H were operating within a classic shifting-gene-freqs paradigm of evo from standing var, where origination events aren't counted, only mass pressures
50 yrs later, Kimura and others developed an origin-fixation paradigm familiar to you. They were mostly interested in neutrality
Note that scientists working within a paradigm don't test its assumptions (and often don't know them). They just use the paradigm.
2/
You use an OF paradigm that gives one answer. The SGF paradigm gives a different answer
To probe the issue, one must step outside of both. Joanna and other scientists like myself have been doing this for 20 years and our work suggests some answers
If the alleles relevant to the outcome of evo are present initially, typically the best allele wins and u isn't important. This is the buffet regime of SGF dynamics, where selection dominates (in large pops).
3/
The other extreme is the sushi conveyor regime of OF dynamics when uN << 1
As uN increases, we depart from OF regime and see clonal interference, when multiple beneficial variants compete, slowing down adaptation
In finite problems, high uN eventually drowns out effects of arrival bias. But in infinite spaces, clonal interference slows down adaptation but does not kill effects of arrival bias bc there are always mut-favored variants in the most-fit class
4/
This is based on Gomez, et al 2020 and Cano, et al 2022, 2023
That's theory. Cano, et al 2023 also review empirical support
I don't find it helpful to say this is implicit in Kimura. Many people have worked hard to explore this issue theoretically and find empirical support
The notion that arrival biases are important in evo is not a truism, and IMHO it is not a hidden implication of well established knowledge, but a provocative and testable conjecture
5/5
Svensson (2023) is in the recently released Dickins & Dickins volume that you can find online here:
Alas, the lack of review is not the issue: 2019 TREE, 2022 & 2023 were all reviewed in some way
The issue is reviewers and editors who do not know this new line of arg, and are taking their cues from Svensson, who makes the work sound flakey, repeatedly misdefining the theory and linking it with directed mut
e.g., in this 2019 fig, the arrow with "?" represents a theory of mut bias as an "independent cause" of adaptation. Not our language, not our figure, but a lie
e.g., here is 2023 @EvolOdonata putting "directed mut" in the same section title as "mut-driven evo" (a totally different thing), and then loading up his rhetorical cannon with strawman args to fire at us
@arlin @JoannaMasel @EvolOdonata Okay I found that in preprint form, I think: https://ecoevorxiv.org/repository/view/4002/
Found this in a section called "5.7 The reemergence of mutation-driven evolution and directed variation?":
1/
@arlin @JoannaMasel @EvolOdonata "Given the strong experimental and empirical evidence against directed mutations (Lenski and Mittler 1993; Futuyma 2017; Svensson and Berger 2019) and the failure of the early mutationists to appreciate the power of natural selection, it is astonishing that some contemporary evolutionary biologists are pushing for a revival of mutationism or mutation-
driven evolution (Stoltzfus 2006; Nei 2013; Stoltzfus and Cable 2014)."
2/
@arlin @JoannaMasel @EvolOdonata I don't know your work super well but I suspect that passage is misleading at best, especially under a heading like that.
I'd love to be wrong.
3/3
@JoannaMasel @arlin @EvolOdonata I bought the book! Since we just moved to a new place, it is still unwrapped 😑. Sadly for me, besides being busy with a move, I'm blogging through a truly awful book by Simon Conway Morris. As ELO once sang, "A fool and his money soon go separate ways." Heh. When I'm done with that, I'll unwrap Arlin's opus.
Unfortunately it's easy to tell that Svensson is misrepresenting other people's work. It's harder to tell whether this is intentional.
Yes, like @JoannaMasel I don't like "mut-driven"
Ppl use this with an explanatory meaning but in #popgen it connotes a causal concept of mass-action forcing
@EvolOdonata turns this into a classic bad-faith arg: ppl who refer to "mut-driven" evo are mistaken, he says, bc selection (not mutation) is driving alleles to fixation (but that's not what anyone means!)
BTW, I cite the passage you quote to show how cultural bogeymen are used to shut down dialog (https://www.molevol.org/conceptual-immune-system/)
Arlin Stoltzfus
Julien JOSEPH
Joanna Masel
Dataviking 🦣
Stephen Matheson🌵🌲