Is the neutral theory dead?

Motoo Kimura. Source: WikimediaCommons

You might have noticed how the world of genetics was shaking as the giants of theoretical population genetics started discussing some of the most fundamental questions in the arena of Twittersphere. This happened after the publication of Andrew Kern and Matthew Hahn’s paper titled “The Neutral Theory in Light of Natural Selection“.

For the sake of all the early career scientists who might be under the false impression that they don’t understand anything, while all the fancy last authors have it figured out, I have decided that at least some of the Twitter gems shall be recorded.

I’m not going to walk into the lion’s den and try to argue for or against neutral theory or even comment on the new paper. I’m merely providing a record of what’s been said by others, and by the end of this post you might be more confused about neutral theory than ever (#sorrynotsorry).

I think this could be the best summary of the whole thing:

To briefly explain what the (by some called controversial) paper is about, let me cite the abstract itself: “We argue that the neutral theory was supported by unreliable theoretical and empirical evidence from the beginning, and that in light of modern, genome-scale data, we can firmly reject its universality.”

Although many people did not see that coming, already in 1996 Martin Kreitman published a paper titled “The neutral theory is dead. Long live the neutral theory“, where he wrote: “…the neutral theory would eventually be shown not to be adequate as an explanation of many features of molecular variation and evolution, while, at the same time, would provide the very framework for proving this thesis. This point of view explains my dependence on this easily falsifiable theory and my desire for its longevity.”


And Twitter responded in all sorts of ways.


Kern and Hahn’s paper polarised the popgen Twittersphere.


The whole debate goes as deep as to the question of how much genomic variation is shaped by natural selection and what are the relative roles of positive and negative selection in shaping the patterns of variation.

In the new paper, Kern and Hahn (2018) lay out the original lines of evidence for the neutral theory and step by step explain why they are not only wrong, but why they were never correct in the first place. For instance, based on Haldane’s estimate of ‘the cost of natural selection‘ (Haldane 1957), Kimura argued that the rate of evolution is too high to be explained by selection and many mutations must be neutral (Kimura 1968). However, Kimura’s estimate was immediately challenged, e.g. by Maynard Smith (1968), who suggested that the rate of evolution reported by Kimura might be overestimated by several degrees of magnitude.

From large datasets and genomic data it is now clearer that selection has far-reaching consequences for genomic variation at linked neutral sites through hitch-hiking (Maynard Smith and Haigh 1974) and background selection (Charlesworth et al. 1993). Some of the most striking imprints of selection can be seen in the correlation between recombination rates and levels of polymorphism (Begun and Aquadro 1992) and between recombination rates and allele frequencies (Stajich and Hahn 2005; Lohmueller et al. 2011).

“We document several significant correlations between different genomic features. In particular, we find that average minor allele frequency and diversity are reduced in regions of low recombination and that human diversity, human-chimp divergence, and average minor allele frequency are reduced near genes. […] Further, we show that the widespread presence of weakly deleterious alleles, rather than a small number of strongly positively selected mutations, is responsible for the correlation between neutral genetic diversity and recombination rate.” (Lohmueller et al. 2011)

Average minor allele frequency versus recombination rate. Source: Fig. 1B in Lohmueller et al. 2011

But the major subject in the debate that emerged after the publication of Kern and Hahn’s paper is whether we should abandon the neutral model completely or not. Analyses of genetic variation assuming that all SNPs are selectively neutral can be misleading if many of these are actually linked to sites under selection and the frequency spectrum used for demographic analyses is skewed by SNPs originating from regions near genes and with low recombination rate (Lohmueller et al. 2011).

“Rejecting the neutral theory does not mean embracing adaptive storytelling, nor does it mean that we must forsake all models that assume neutrality. But we must recognize that assuming a neutral model for the sake of statistical convenience can positively mislead our inferences. One area where this problem is especially dire is in the estimation of demographic histories. While most populations almost certainly have a nonequilibrium history, attempting to infer the details of these histories without accounting for selective forces can mislead us in multiple ways.” (Kern and Hahn 2018)

Regarding the abuse of terminology concerning the neutral theory, there’s a very enlightening post by Casey Bergman where he explains what he titled the “Neutral Sequence Fallacy”, i.e. “the incorrect notion that if a sequence is neutrally evolving, it implies a lack of functional constraint operating on that sequence, and vice versa.

So eventually it boils down to whether the neutral theory should officially be pronounced utterly dead or if it still has its place as formulated by Kreitman more than two decades ago (Kreitman 1996): “It is my thesis that the neutral theory (and noncoding variation) will continue to play a leading role in this quest to detect selection, even as it is being rejected. For this reason, I say, long live the neutral theory.”






If you are interested, at this point there’s a 50-tweet discussion about effective population size, but I’m not going to put it here because that would be a little too much.

Instead, I’ll just put one more interesting discussion.


And I think there’s no better way to finish this post then with a conclusion made by Matthew Hahn himself on Twitter:

Read more on the topic:

Begun DJ, Aquadro CF (1992) Levels of naturally occurring DNA polymorphism correlate with recombination rates in Drosophila melanogaster. Nature, 356: 519–520. doi:10.1038/356519a0

Charlesworth B, Morgan MT, Charlesworth D (1993) The effect of deleterious mutations on neutral molecular variation. Genetics, 134: 1289–1303.

Haldane JBS (1957) The cost of natural selection. J Genet, 55: 511-524.

Kern AD, Hahn MW (2018) The Neutral Theory in Light of Natural Selection. Molecular Biology and Evolution, 35(6): 1366–1371.

Kimura M (1968) Evolutionary rate at the molecular level. Nature, 217: 624-626. doi:10.1038/217624a0

Kreitman M (1996) The natural theory is dead: Long live the neutral theory. BioEssays 18: 678-683.

Lohmueller KE, Albrechtsen A, Li Y, Kim SY, Korneliussen T, et al. (2011) Natural Selection Affects Multiple Aspects of Genetic Variation at Putatively Neutral Sites across the Human Genome. PLOS Genetics, 7(10): e1002326.

Maynard Smith J (1968) “Haldane’s dilemma” and the rate of evolution. Nature, 219: 1114-1116. doi:10.1038/2191114a0

Maynard Smith J, Haigh J (1974) The hitch-hiking effect of a favourable gene. Genetics Research, 23: 23-35. doi:10.1017/S0016672308009579

Stajich JE, Hahn MW (2005) Disentangling the Effects of Demography and Selection in Human History. Molecular Biology and Evolution, 22: 63-73.


About Patrícia Pečnerová

I'm currently a postdoctoral researcher at the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany. My research interests focus on evolutionary processes in extinct and endangered species. For my PhD, I studied genetic extinction factors in the last population of the woolly mammoth. Now I'm working on genomics of wild mountain gorillas
This entry was posted in evolution, mutation, population genetics, selection, theory and tagged , . Bookmark the permalink.