Male guppies have complex, highly-variable, genetically-determined colour patterns, while females have uniform, inconspicuous colouration. In previous experiments, Hughes and Rodd found that male colour polymorphism is too high to be explained by the “classical” neutral model (recurrent deleterious mutation balanced by selection to eliminate variation), thus implicating balancing selection in maintaining polymorphism. Negative frequency dependent selection (NFDS), one form of balancing selection was tested by manipulating colour pattern frequencies in three natural populations and measuring individual survival and reproductive success. In these short-term experiments, males with colour patterns that were locally rare had a significant reproductive advantage, acquiring more than twice as many mates and siring twice as many offspring compared to males bearing common colour patterns.

 

Therefore, we hypothesise that female guppies exhibit a strong mate preference for males bearing rare or unfamiliar colour patterns and performed a series of analyses to test these predictions.

​

Part 1: 

Since guppy colour patterns are often inherited faithfully from fathers to sons, it has been presumed that colour genes are physically linked to sex determining loci as a ‘supergene’ on the sex chromosome. Yet the actual identity and genomic location of the colour pattern genes has remained elusive. We phenotyped and genotyped four guppy ‘Iso-Y lines’, where colour was inherited along the patriline, but backcrossed into the stock population every generation for 40 generations. Using an unbiased phenotyping method to proportion colour pattern differences between and among the Iso-Y lines, we confirmed that the breeding design was successful in producing four distinct colour patterns. Our analysis of genome resequencing data of the four Iso-Y lines uncovered a surprising genetic architecture for colour pattern polymorphism. Genetic differentiation among Iso-Y lines was repeatedly associated with a large and diverse haplotype (~5Mb) on an autosome (LG1), not the sex chromosome (LG12). Moreover, the LG1 haplotype showed elevated linkage disequilibrium and exhibited evidence of sex-specific diversity when we examined whole-genome sequencing data of the natural source population. We hypothesise that colour pattern polymorphism is driven by Y-autosome epistasis.

​

Part 2: 

Exploring the genetic architecture of balancing selection allows us to ask fundamental questions about its power and limitations in maintaining genetic diversity. For instance, how repeatable is balancing selection across different populations? Are certain genes more likely to respond to this type of selection? How does the timescale of balancing selection influence its pervasiveness? What are the effects of past population demography? Moreover, how does balancing selection operate when populations lack starting genetic diversity due to founding events? 

 

Guppies in Northern Trinidad have well-characterised population structure, where mountain ranges and waterfall barriers have led to upriver populations being founded by a small number of individuals, with minimal ongoing migration. Finally, there has been a series of experimental translocations, where guppies have been transplanted into previously guppy-free habitats with new selection regimes. 

 

For this study, we used genome scan approaches to identify regions with signatures of balancing selection in 11 populations of guppies. These populations are those where we have evidence of NFDS operating and include two translocated populations, and four bottlenecked populations.