Neutralism and Selectionism

    Ever since Kimura’s bold publication, which suggested that many of evolution’s innovations were more the product of neutral (or ‘nearly neutral’) mutations driven to fixation within a population through the action of genetic drift and its refinement, molecular and evolutionary biologists have had trouble agreeing upon whether natural selection or neutral drift-based evolution were the main motivators behind phenotypic evolution.  Today, it would seem, we are hardly closer to closing the book on this debate than before.  

    As advanced methods for probing the molecular and genomic frameworks of phenotypic evolution are developed, findings seem to, at least on the surface, weaken the case for neutralism.  For example, the McDonald-Kreitman tests provides evidence that between 30 and more than 90 percent off nucleotide changes in the Drosophila genus and in other organisms go to fixation because they are beneficial.  Furthermore, recent genome-scale data seems to support selectionism (ie. the position that the majority of phenotypic evolution is driven by the fixation of beneficial alleles through the act of natural selection) when considering correlations between within-species allele polymorphisms and between species gene polymorphisms.  

    Nevertheless, Andreas Wagner (2008) suggests that, instead of parsing evolutionary understanding into spectral poles (of likely the same underlying mechanism) of ‘selectionism’ vs ‘neutralism’ we should instead view them as complementary mechanics that operate at different strengths at different times.  Selection acts to fix beneficial mutations throughout a population, often acting in large ‘selective sweeps’ that show punctuated periods of adaptation followed by periods of seemingly static existence.  Wagner argues that during such periods of stasis (or even throughout the entire course of evolution), neutralism serves as a massively dynamic force in adaptation, allowing the exploration of so-called “neutral networks”, wherein all possible phenotypic/molecular possibilities are connected, and SNPs of near-neutrality represent explorations of such a network.  This standing genetic variation within a population actually acts to “feel out” all the nearly-neutral phenotypic possibilities, sometimes acting to ‘discover’ adaptive innovations that, under a strictly selectionist regime, would be impossible.  

    Consider a molecular phenotype within a population.  Within this functional molecule there are many possible mutations that would be neutral, or, at least, nearly neutral, in that they wouldn’t drastically alter the molecule’s function.  Through neutral mechanisms (neutral SNPs that alter the underlying genetic code, or nearly neutral amino acid substitutions that do not noticeably alter the proteins function) large amounts of variation can be accrued over time within the population.  As the molecule fills its neutral “coding space”, it is brought closer to possibly beneficial mutations than could have ever been possible by considering only beneficial mutations.  Furthermore, neutrality’s power can be even more pronounced when one considers the often quixotic nature of selection in determining what is and is not a “fit” phenotype (or fit character).  In some instances, mildly deleterious mutations might be allowed within a population due to the fact that the protein in question is not the primary predictor of fitness at that time (eg: being able to run from a lion is no longer the primary predictor of human fitness, and therefore less defining in terms of a given human's survivability).  These mechanisms and others allow for evolutionary spurts, or punctuated innovation where a population undergoes molecular “exploration” at a specific loci (or many loci), followed by selective sweeps once a adaptive innovation is “discovered”.

    Though pedestrian, this explanation is largely what lies at the root of Dr. Wagner’s contentions in support of a unified view of evolution.  He argues that neutral and selective “regimes” dominate at different times, likely as the product of these different mechanisms, allowing for evolutionary innovation that would be otherwise impossible under strictly selectionist or strictly neutralist considerations.

-Fay, J.C. (2002) Testing the neutral theory of molecular evolution with genomic data from Drosophila DNA. Nature 415: 1024-1026

-Wagner (2008) Neutralism and selectionism: a network-based reconciliation. Nature Rev. Genetics 9: 965-974