January 05, 2014

Population size and the rate of evolution (Lanfear et al. 2014)

A useful treatment of a very general subject. From the paper:
For mutations on which natural selection can act (i.e., those
with s != 0, Box 2), the NeRR depends on the fitness effects of
mutations (s, Figure 1). As Ne increases, natural selection
becomes more effective at fixing advantageous mutations
and removing deleterious mutations, but larger populations
also produce more of both types of mutation. Theory sug-
gests that as Ne increases the power of natural selection
increases faster than the production of new mutations (see
[5] for a recent review). This results in lower deleterious
substitution rates as Ne increases (a negative NeRR,
Figure 1B,D), and higher advantageous substitution rates
as Ne increases (a positive NeRR, Figure 1A,C).
However,
these predictions can sometimes be altered when the sim-
plifying assumptions of the underlying theory are not met.

Trends in Ecology & Evolution Volume 29, Issue 1, January 2014, Pages 33–41

Population size and the rate of evolution

Robert Lanfear et al.

Does evolution proceed faster in larger or smaller populations? The relationship between effective population size (Ne) and the rate of evolution has consequences for our ability to understand and interpret genomic variation, and is central to many aspects of evolution and ecology. Many factors affect the relationship between Ne and the rate of evolution, and recent theoretical and empirical studies have shown some surprising and sometimes counterintuitive results. Some mechanisms tend to make the relationship positive, others negative, and they can act simultaneously. The relationship also depends on whether one is interested in the rate of neutral, adaptive, or deleterious evolution. Here, we synthesize theoretical and empirical approaches to understanding the relationship and highlight areas that remain poorly understood.

Link

6 comments:

barakobama said...

I would assume evolution occurs faster in larger populations, because there is a better chance a new mutation that helps survival will appear.

Grey said...

I wonder if one consequence of this might be dating by number of mutations? If larger populations have a faster rate of mutation than smaller ones then that might distort the estimated age.

For example a population with a particular y marker somewhere warm and high density might generate more mutations per century than a population with a similar y marker who lived somewhere cold and low density.

Anders Pålsen said...

Maybe substitutions get confused as recombination in larger and more diverse populations than among smaller and less diverse populations?

apostateimpressions said...

The natural selection of adaptations occurs when people without those adaptations either rapidly or gradually die off. Selection entails death. Selection does not occur when populations explode because then everyone survives and breeds and there is no adaptation without selection.

Evolution works best on a small population with harsh conditions because new adaptations then spread rapidly through the population.

Evolution would also work on a large population that became subject to demanding conditions and which then had a high death rate and became a smaller population. Death is key to evolution.

Large populations without selection will accumulate junk and deleterious mutations and they will lose adaptations. Species and population groups lose adaptations when those adaptations no longer provide a survival benefit.

Brad Foley said...

@apostateimpressions: False. Natural selection occurs because of differential reproduction, and does not necessarily entail death. Different alleles may be selected during invasion events and massive population explosion (e.g.), due to low competition and fecundity selection than are selected during stationary population sizes (think a now-classic model in cane toads).

@barakobama: selective sweeps on low-effect size favourable alleles happen more effectively in large populations (there were a couple of graphs of this process in the paper). Drift is also evolution, and happens faster in small populations. I was surprised (though I guess I shouldn't have been) that it seems as if purifying selection pervasively *slows down* the rate of evolution in large populations, due to more efficient selection on mildly deleterious alleles.

@Grey: there are a number of demographic models that use haplotype block size, accumulation of mutations in sweep regions, dN/dS ratios etc etc ad nauseum to estimate exactly the kind of demographic parameters (effective population size, expansion, bottlenecks and things) you're talking about here. I guess looking at branch lengths from a common population origin might get at the differential rate question? At some point, it's difficult to untangle a few parameters (mutation rate and Ne notoriously).

Unknown said...

Brad Foley said...
‘Natural selection occurs because of differential reproduction...”

That can’t be right, except in very narrow genomic modeling. Obviously, natural selection is more than just a reproduction contest.

When new traits arise, they don’t need to compete for natural selection to happen.

For example, there are birds on an island who eat nuts. They randomly mutate into birds who eat fruit. The reproduction rate of either bird population is independent of the other. They don’t compete for food. Differential reproduction does not matter, but there is natural selection. Both populations of birds are fit and both are positively selected.

To say that the two kinds of birds have become separate populations -- which is the pat answer -- is to try to force natural selection into always being a competition. The genomic model does not properly account for diversity.

Another problem here is what does “natural” selection mean, particularly in a human context.

For example, it becomes the fashion in a closed group of humans that long eyelashes are in. No one will mate or marry with men or women who have short eyelashes. In a few generations, short eye lashes completely disappear from this group.

Is this natural selection or what Darwin called “artificial selection”?
 What made long eyelashes “advantageous” or short ones “deleterious?”
Are these kinds of changes subject to any of the projections of genomic modeling? The answer is in most cases, no.

Finally, rate of biological evolution is going to depend, first of all, on rate of mutation. The more complex an organism becomes, the fewer pathways there are for viable mutation. The trouble with genomic modeling in the article above is that it does not account for this obvious constraint on evolution. And rate of mutation an organism must attain to keep pace, like the Red Queen, with an ever-changing environment.
s. long