Fruit Flies Not Evolving

A long-running experiment trying to get fruit flies to evolve has failed.  A research team forced selection on the flies to explore the limits of natural selection.  Only minor changes were detected after 600 generations.  The research team was disappointed and surprised; there was even less evolution in these sexual organisms than in similar experiments with microbes, like bacteria and yeast (but see 07/12/2010).  And all this was under ideal lab conditions.  Success is even less likely in the wild.
    The Editor’s summary of a paper in Nature was titled, “Experimental evolution reveals resistance to change” and ended that the authors “conclude that unconditionally advantageous alleles rarely arise, are associated with small net fitness gains, or cannot fix because selection coefficients change over time.”  Nature this week published the results of a 35-year study by UC Irvine and University of Southern California (USC).  Here is the abstract:¹ 

Experimental evolution systems allow the genomic study of adaptation, and so far this has been done primarily in asexual systems with small genomes, such as bacteria and yeast.  Here we present whole-genome resequencing data from Drosophila melanogaster populations that have experienced over 600 generations of laboratory selection for accelerated development.  Flies in these selected populations develop from egg to adult ~20% faster than flies of ancestral control populations, and have evolved a number of other correlated phenotypes.  On the basis of 688,520 intermediate-frequency, high-quality single nucleotide polymorphisms, we identify several dozen genomic regions that show strong allele frequency differentiation between a pooled sample of five replicate populations selected for accelerated development and pooled controls.  On the basis of resequencing data from a single replicate population with accelerated development, as well as single nucleotide polymorphism data from individual flies from each replicate population, we infer little allele frequency differentiation between replicate populations within a selection treatment.  Signatures of selection are qualitatively different than what has been observed in asexual species; in our sexual populations, adaptation is not associated with ‘classic’ sweeps whereby newly arising, unconditionally advantageous mutations become fixed.  More parsimonious explanations include ‘incomplete’ sweep models, in which mutations have not had enough time to fix, and ‘soft’ sweep models, in which selection acts on pre-existing, common genetic variants.  We conclude that, at least for life history characters such as development time, unconditionally advantageous alleles rarely arise, are associated with small net fitness gains or cannot fix because selection coefficients change over time.

In other words, they looked for evidence of a “selective sweep” – the signature of a beneficial mutation becoming fixed in the population – and could not find it.  They did the selection artificially, forcing the fly embryos to evolve toward faster embryonic development.  Despite lots of mutations, they found the flies resistant to change.  Not only that, the flies underwent “reverse evolution” – they said, “forward experimental evolution can often be completely reversed with these populations, which suggests that any soft sweeps in our experiment are incomplete and/or of small effect” (a soft sweep meaning selection is acting on standing variation instead of new mutations).  Possibly any beneficial mutations were hindered by linked deleterious alleles (canceling out the benefit) or antagonistic pleiotropy (in which one good mutation to a gene can cause one or more bad effects elsewhere).  Either way, the evolution is like one step forward, one or more steps back.
    There was even more bad news for neo-Darwinian theory: the lab situation was more optimistic than the wild, where adaptive evolution is expected to occur.  You can get a lot of variation and mutation to appear in genomes, but no unconditionally beneficial mutations.  Their last paragraph expressed surprise at this, with a subtext of disappointment:

Our work provides a new perspective on the genetic basis of adaptation.  Despite decades of sustained selection in relatively small, sexually reproducing laboratory populations, selection did not lead to the fixation of newly arising unconditionally advantageous alleles.  This is notable because in wild populations we expect the strength of natural selection to be less intense and the environment unlikely to remain constant for ~600 generations.  Consequently, the probability of fixation in wild populations should be even lower than its likelihood in these experiments.  This suggests that selection does not readily expunge genetic variation in sexual populations, a finding which in turn should motivate efforts to discover why this is seemingly the case.

This experiment was begun in 1975.  After 35 years and 600 generations, accelerated by artificial selection, the net evolution (in terms of adaptation and improvement in fitness) was negligible if not nil.

1.  Burke, Dunham et al, “Genome-wide analysis of a long-term evolution experiment with Drosophila,” Nature 467, 587-590 (30 September 2010); doi:10.1038/nature09352.

Natural selection is always presumed to be the wonder-worker that can produce eyes, ears, sonar, flippers, jaws, hearts, and brains with its gradual, step-by-step improvement of natural variation, without design (07/20/2010, 05/04/2010).  OK, where is it?  It doesn’t work theoretically (09/28/2010, 06/11/2010, 03/21/2010, 03/17/2003), it doesn’t work rhetorically (04/17/2010), it doesn’t work historically (08/05/2010), and it doesn’t work experimentally (09/22/2010).  It doesn’t work in the lab, and it works less in the wild.  Unless you include “reverse evolution,” (06/26/2010), it doesn’t work at all.  Game over, Charlie (05/14/2010).  Stop the hype (08/13/2010).