Friday, August 17, 2007

The Cause of Variation in a Population

John Dennehy of The Evilutionary Biologist has posted a wonderful article on This Week's Citation Classic. The classics are two back-to-back papers on genetic variation in fruit flies (Hubby and Lewontin (1966), Lewontin and Hubby (1966)). That's Lewontin on the left.

Please get on over to The Evilutionary Biologist and read what John has to say. These were very important and groundbreaking papers when they came out and everyone needs to know why.

Here's some background.

In the olden days there were two competing theories to explain variation (heterozygosity) in a population. The classical theory said that mutations are constantly being removed from the population by positive natural selection or purifying selection. Variation is a transient phenomenon that would disappear entirely if it weren’t for new mutations that arise at a significant rate.

The balance theory maintains that variation in a population is often due to balancing selection. The best known example of balancing selection is the allele for sickle cell disease. In the heterozygous state it confers resistance to malaria but in the homozygous state it is often lethal. Both the sickle cell allele and the wild type allele are maintained in the human population by balancing selection.

Hubby and Lewontin (1966) discovered that there was a huge amount of genetic variation in fruit flies. Their data suggested that 50% of all loci had multiple alleles. This is difficult to reconcile with the balance theory and it was also a big surprise to those who supported the classic theory. It seemed unlikely that at any given point in the evolutionary history of a species that so many genes could be undergoing selection. Further work confirmed that other species contained a huge amount of variation.

The solution to this surprising observation was the recognition that most of the alleles were neutral. The variation is explained by fact that fixation by random genetic drift is much slower than fixation by natural selection. Thus, while the variation is transient in the sense that it is a snapshot of an ongoing process, the process is not selection but drift.

The results of Hubby and Lewontin (1966) led directly to Neutral Theory.
The neutral theory also asserts that most intraspecific variability at the molecular level (including DNA and protein polymorphisms) is selectively neutral, and is maintained in the species by the balance between mutational input and random extinction. In other words, the neutral theory regards protein and DNA polymorphisms as a transient phase of molecular evolution and rejects the notion that the majority of such polymorphisms are adaptive and actively maintained in the species by some form of balancing selection.
                M. Kimura
This explanation is also known as the Neoclassical Theory. Balancing selection is now thought to play only a minor and insignificant role in the cause of variation in a population.
... the neoclassical theory is not refuted by occasional observations of overdominance for fitness, because the theory does not deny that cases exist but only that they are common and explain a significant proportion of natural variation. So it is no use trotting out that tired old Bucephalus, sickle-cell anemia, as a proof that single-locus heterosis can exist. Anyone who has taught genetics for a number of years is tired of sickle-cell anemia and embarrassed by the fact that it is the only authenticated case of overdominace available. “If balancing selection is so common," the neoclassicists say, "why do you always end up talking about sickle-cell anemia?"
                R. Lewontin

[Photo credit: The photograph of Richard Lewontin is from (Photographs of Participants in the Molecular Evolution Workshop)]


  1. Over at History of recent Science & Technology you will find a video of Lewontin working in the lab back in the 70s. In addition you will find videos with Kimura and Dobzhansky here. All clips are from BBC's "Life Game" that was aired in 1973.

  2. I posted the link to the Lewontin clip in my previous comment to give you an impression of the 70s electrophoresis equipment. If I had watched it before I could have written in my first comment that the clip is indeed on topic.
    Here's what Lewontin says in the video:
    "For a long time, we didn't know how to determine how much genetical variation there was between individuals, even though we could see surface differences between them. But now these new methods of molecular biology, like gel electrophoresis, make it possible to determine, gene by gene, the differences and similarities between individuals in the same population. And when we apply these methods, we find huge variations from individual to individual, not only in Drosophila but in man as well. . . . In this way, molecular biology has solved a longstanding problem of evolution, which is, how much genetical variation is there from individual to individual that makes evolution possible."

  3. Thanks, sparc. That's very much on topic.

  4. Great background, Larry! I didn't really cover the relevance to the Classic/Balance controversy.

  5. When I used to teach the molecular half of the sophomore genetics course at Union College, my senior colleague who taught the classical and population material was a guy who had gotten his Ph.D. in Lewontin's lab at the University of Chicago. We had a lab exercise in the course using one of the classic old Lewontin and Hubby enzyme electrophoresis protocols. Those papers are one of the great moments in the history of genetics, for sure.