Divergence and parallel evolution of boldness in guppies

点击翻译按钮获取中文摘要

The evolutionary origin of variation in animal behaviour continues to both fascinate and frustrate biologists. Despite decades of research, we still lack a comprehensive understanding of the mechanisms that sometimes promote variation among individuals, populations and species, but can also limit diversity and constrain divergence. Here we propose a study of guppies, a small freshwater fish from Trinidad, to investigate how, when and why the structure of behavioural variation changes across populations.Our study is 'comparative' - we will investigate patterns of behavioural variation arising from evolutionary divergence among multiple wild populations of guppies. It will also be 'multi-level' - we seek to understand patterns of similarity and difference among populations, but because natural selection happens among individuals within populations, to achieve this requires studying variation among individuals and genotypes. How two populations that have become separated evolve ('microevolution') and diverge ('macroevolution') does not only depend on natural selection, but also on the way in which genetic factors contribute to behavioural differences among individuals. Moreover, gene frequencies change over time even in the absence of natural selection - so just how much microevolutionary change and macroevolutionary divergence is the result of 'neutral' evolution?We will focus on 'boldness', an aspect of behaviour that can be broadly understood as describing an individual's response to risky conditions in the environment. The set of populations we will study have evolved in environments differing in predation regime, which is thought to be an important source of selection on boldness. We will combine large scale behavioural data collection in the laboratory, with experiments to manipulate perceived predation risk, and both statistical genetic and genomic methods. Comparing the structure of behavioural variation across populations of known evolutionary relationships to each other (or 'phylogeny') will allow us to robustly test hypotheses for the evolutionary factors maintaining shy-bold variation within populations, and driving differences among populations. By combining uniquely comprehensive genomic and behavioural data, we will test the extent to which differences in behaviour between populations arise from selection versus 'neutral' evolution of the underlying genetics that is expected even if behaviours are not under selection. They will also let us probe the genetic 'building blocks' of behavioural variation, providing valuable new insights into the links between genotype and behaviour.