Molecular drive is an evolutionary process, like natural selection and driftปกติ, that changes the genetic composition of a population, through the generations. It is distinct from natural selection and neutral drift in that it emerges from the activities of a number of ubiquitous mechanisms of DNA turnover (MOT), such as gene conversion, unequal crossing over, slippage, transposition, retrotransposition and so on.
So, how does it work?
Consider a single mutation arising at a single location, on a single chromosome, in a single individual. The theories of natural selection and neutral drift assume that this mutation cannot increase in frequency in a population under its own steam because the chromosome on which it resides obeys Mendel's rules of inheritance, which in turn depend on the random assortment of chromosomes at meiosis. By contrast, all MOT are essentially non-mendelian, in that the initial mutant sequence can increase or decrease in copy number within the lifetime of an individual. Take, for example, gene conversion, whereby one allelic version of a gene is converted to the sequence of another version on the homologous chromosome, such that Aabecomes AA or aa.
How does this translate into molecular drive?
The two homologous chromosomes, now carrying alleles AA or aa (rather than Aa), will inhabit two different individuals at the next generation, in both of which gene conversion could re-occur. In this way A or a may spread. If gene conversion is biased in favour of say A, then A will spread that much faster. The other MOT, although different in kind, have essentially the same consequence. Unstable DNA has a lot to answer for.
But isn't all gene segregation mendelian?
Yes and no. The rates of MOT (∼10−2 to 10−4 per generation per gene) are too slow to affect mendelian ratios when small numbers of progeny are examined, often between a few generations only. However, among thousands of progeny, or over thousands of generations, deviations from mendelian expectations can become significant. Chromosomes are mendelian; DNA is not, on an evolutionary timescale. This crucial separation between DNA behaviour (due to MOT) and chromosome behaviour (due to meiosis) is at the heart of molecular drive.
What is the evidence for molecular drive?
There are three lines of evidence. First, the ubiquitous existence of MOT. Few stretches of plant, animal, microbial or viral DNA, whether a gene, a regulatory region or ‘junk’ DNA are refractory to one or other MOT, which often operate one on top of another, with different rates, units and biases. Second, the widespread observation of ‘concerted evolution’. This is where multiple copies of a given DNA sequence (which could be a gene or non-genic DNA) share an identical mutation that is specific for a given species, implying the mutation occurred once in the species and spread through most copies in most individuals. When concerted evolution embraces families of DNA repeats consisting of thousands of genic or non-genic members, scattered over several pairs of chromosomes, it is difficult to offer a convincing explanation based on natural selection and/or neutral drift acting in the absence of other processes. However, with the discovery of the spreading consequences of MOT, concerted evolution can be easily accommodated.
Molecular drive is the population genetics process that explains how such species-specific patterns of multiple-copy homogeneity arose during evolution. Essentially, molecular drive can be considered a two-step process: the gradual spread of a sequence variant (1) through a genome (homogenization), and (2) through a population (fixation) (Fig. 1). Because MOT are roughly two orders of magnitude faster than the basic mutation rate (10−6 per gene per generation), yet very much slower than the rate at which the sexual process scrambles chromosomes among individuals at every generation, homogenization and fixation go hand in hand. If, for example, a new mutation is half-way to being homogenized in a bunch of repetitive elements, then in a sexual population no individual remains unhomogenized and no individual races ahead to becoming fully homogenized. The cohesive manner in which a population changes genetically under molecular drive (like ripples in a pool) has important consequences for the interaction with natural selection, and for any new relationship between the population and its environment.