Academic journal article Genetics

An Age-of-Allele Test of Neutrality for Transposable Element Insertions

Academic journal article Genetics

An Age-of-Allele Test of Neutrality for Transposable Element Insertions

Article excerpt

(ProQuest: ... denotes formulae omitted.)

NATURAL selection against transposable element (TE) insertions is considered to be one of the primary forces preventing their proliferation in populations. The action of negative selection against these genetic parasites is thought to come in three predominant forms: selection against insertions in functional regions (Charlesworth and Langley 1989), chromosomal abnormalities arising from ectopic recombination (Montgomery et al. 1987; Langley et al. 1988), and costs associated with the transposition process itself (Nuzhdin et al. 1996). Understanding the relative importance of each of these forces has been of substantial interest for many years (Charlesworth and Langley 1989; Charlesworth et al. 1994; Nuzhdin 1999; Lee and Langley 2010). To understand the nature of selection acting on TEs, a common practice is to measure the allele frequency distribution of TE insertions within natural populations (Montgomery et al. 1987; Biemont et al. 1994; Petrov et al. 2003, 2011; Yang and Nuzhdin 2003; Gonzalez et al. 2008; Kofler et al. 2012). These studies have found that TE insertion alleles segregate at low allele frequencies in Drosophila melanogaster, and this observation has been used to support the idea that negative selection acts to prevent TE insertions from increasing in frequency in populations (Charlesworth and Langley 1989).

A limitation of previous studies on the dynamics of TE evolution is that the frequency distribution under different models of selection is typically evaluated under the assumption of transposition-selection balance within the population (Charlesworth and Langley 1989; Petrov et al. 2003; Lockton et al. 2008; Gonzalez et al. 2009; Lee and Langley 2010). A crucial assumption of models that posit transposition- selection balance is that the transposition process can be modeled as a constant rate over time. This is often unlikely to be the case, as episodes of transposition are known to occur in bursts. For example, the P element invaded and proliferated in D. melanogaster only within the past several decades (Kidwell 1983; Daniels et al. 1990). Likewise, analysis of genome sequences has demonstrated waves of transposition for a number of other TE families (Sanmiguel et al. 1998; Lander et al. 2001; De La Chaux and Wagner 2009; Lu et al. 2012). In cases of recent transposition bursts, insertion allele frequencies will not be at equilibrium because there will not have been sufficient time to driftto moderate or high allele frequencies, even under strict neutrality. Therefore recent insertion alone may explain the pattern of low allele frequencies for TE insertions observed in natural populations of D. melanogaster (Bergman and Bensasson 2007). Alternatively, negative selection may explain the pattern since equilibrium can be achieved quickly when TEs are harmful. To distinguish among these possibilities, it would be beneficial to relax the assumption of transposition-selection balance in models of TE evolution. We develop such an approach here. To relax equilibrium assumptions we ask the following: Are TE insertion allele frequencies consistent with neutrality, conditional on the inferred time that has elapsed since insertion? If so, then one may conclude that genetic driftand demography are the major factors shaping the evolution of TE insertion allele frequencies. However, if TE insertions are observed at a lower frequency than predicted based on their age, we may infer that negative selection is limiting their increase. Alternately, if a TE insertion is at a higher frequency than expected based on its age, we may infer the action of positive selection on that allele.

Critical to this approach is being able to estimate the time that has elapsed since origination of the insertion allele. For most mutations, information about allele age is provided solely by the frequency of the allele itself or in the amount of linked variation (Slatkin 2000). Under neutrality, a low-frequency allele is on average younger than a high-frequency allele (Kimura and Ohta 1973) and alleles with low levels of linked variation and greater haplotype structure tend to be younger because there has not been sufficient time to accumulate mutations or undergo recombination (Slatkin 2000). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.