A little bit of sex goes a long way

By Matthew Hartfield, EEB. Matty Hartfield gives us a digestible discussion of his latest paper, coauthored with Stephen Wright and Aneil Agrawal, which is available on biorxiv.

Most evolutionary biologists are obsessed with sex. Explaining why organisms reproduce sexually by combining genetic material is a tough problem. The main issue lies with the fact that asexuality (reproduction via clonality) should be able to outcompete sexuals due to sheer weight of numbers. Various theories have been put forward to explain why sex is so widespread. The majority of these revolve around the idea that exchanging genetic material enables the fittest possible genotype to be created, while that of asexuals should degrade over time.

While such theories are ubiquitous, data to test them has been scarce. Recent years have seen a boom in exploring the evolution of sex experimentally using facultative sexual organisms: species that can switch between sexual and asexual reproduction. Such experiments have demonstrated how sexual reproduction can evolve when exposed to stressful environments, or when moving between environmentally different areas. Yet major questions remain regarding what the underlying genetic causes of these transitions are. In addition, there are plenty of organisms that undergo ‘cryptic’ sex, which cannot be observed directly but can with genomic sequence analyses.

Coalescent models are important for analysing genomic data. These tools determine the relationship between neutral markers, and hence make predictions on how genetic diversity is affected depending on environmental structuring, localised natural selection, or other effects. However, classic models cannot be applied to systems with partial asexuality, as they assume the population reproduces entirely sexually.

We worked on introducing partial rates of sex into these models. In the simplest case (one population with a fixed rate of sex), we recovered a classic prediction that extensive divergence between alleles at the same site arises. This phenomenon occurs since lack of sex keeps the two alleles distinct over evolutionary time; only a rare bout of sex has any chance of creating the segregation needed for them to be descended from the same allele.

A schematic of Allelic Sequence Divergence (ASD) in asexuals: Xs are distinct mutations at each neutral site.

After recovering this familiar result, we worked to extend coalescent theory in partial asexuals to include various other biological phenomena. Two effects we looked at were gene conversion, and heterogeneity in sex rates that change over time or space.

Gene conversion, where one DNA sequence replaces part of a homologous chromosome, is usually regarded as being of minor evolutionary importance. Yet numerous studies of facultative sexuals often observe it as a common force, especially in species not exhibiting allelic sequence divergence (ASD). Could the two be related? Excitingly, we found that low rates of gene conversion become important in organisms with low rates of sex. That is, once sex becomes so rare as to caused ASD, small rates of gene conversion can then reverse the process, homogenizing alleles again. Rather than having higher diversity than otherwise similar sexual populations as expected with ASD (in the absence of gene conversion), asexual populations will have less diversity than comparable sexual populations if gene conversion is not too low.

It is also known that many organisms change their rates of sex over time or location, which can be triggered by environmental cues or organismal stress. By investigating such variation in the rate of sex, the analysis elegantly shows how even a short burst to obligate sex (over tens of generations) is enough to jumble genomes in the population, hence giving the same outcome as long-term obligate sex. If rates of sex are also different in separate geographical locals, then these differences can be detected if there is little gene flow between regions. Otherwise, both areas display intermediate rates of sex.

Coalescent tools are popular since they can be used to simulate complex evolutionary outcomes, which are then tested against genomic data. We used the mathematical analyses to outline a coalescent algorithm to account for partial rates of sex, and predict genetic diversity, under numerous scenarios. The code is available online (http://github.com/MattHartfield/FacSexCoalescent) for others to use.

These are exciting times for population genetics and evolution, with cheaper sequencing costs making it possible to wade through the genomes of more individuals than before. Yet accurately exploring the genetic landscape requires the creation of mathematical tools that accounts for organismal life history. These results will provide the first of many buildings blocks to determine the effects of selection and the environment on the evolution of facultative sexuals. They might eventually reveal why sex is so prevalent in nature.

:: Read the original here ::