Genomics of adaptation in experimental populations of Trinidadian guppies (Poecilia reticulata)

Abstract

Natural populations are increasingly affected by human-mediated changes in the environment. Despite substantial evidence for rapid phenotypic evolution in response to changing environments, the role of genetics underlying rapid adaptation remains relatively unexplored. Natural populations of guppies (Poecilia reticulata) in Trinidad show clear repeatable phenotypic adaptation to low- (LP) and high (HP) -predation environments. Experiments where guppies were transplanted from HP to LP environments showed the LP phenotypes can evolve in as little as four years. Using whole genome sequencing of two well-established introductions (sampled at 64 and 114 generations post introduction), and four newly introduced populations (sampled 8-10 generations post introduction), I investigate the genetic response to novel environments. I uncover varying demographic histories between the two established populations, with evidence of bottlenecks in one and extensive population growth in the other. Both experimental populations showed signatures of convergent evolution with a natural LP population, but I found little evidence of convergence between them, indicating they explored different molecular pathways to achieve similar phenotypes. In the second half of the thesis, I show four recently introduced populations evolved minor genetic changes with respect to their HP source, and find no evidence of bottlenecks in three of the four populations. I find signals of selection in all four populations, and a 2Mb region on chromosome 15 showed a consistent signal of selection in three of the four populations. Finally, using a multivariate analysis of allele frequency changes I uncovered subtle parallel changes at multiple loci across all four populations, indicating the approach has potential to detect convergent evolution in rapidly evolving populations, as well as identifying signatures of polygenic selection. Taken together, the findings in this thesis contribute to a better understanding of the process of rapid genetic adaptation after a sudden environment shift in natural populations.