Social behaviour

Despite the appearance of cooperation in many social systems, natural selection willSlide1 generally favour exploitative individuals who maximize fitness by performing less of a costly cooperative act while coercing other individuals to perform more. The evolution and maintenance of cooperation is therefore counterintuitive as it is characterised by conflict between the self-interests of cooperating individuals. Great strides have been made in addressing this problem using models of social evolution based on the principle of fitness maximisation. However, because these ‘optimality’ approaches often make no explicit genetic assumptions or assume that trait evolution is unconstrained by genetics, they are most successful for predicting possible evolutionary

Slide1outcomes, not evolutionary processes. Consequently, there is growing recognition that understanding the process of social evolution also requires explicit consideration of the underlying genetics of social traits. For example, different traits, social and otherwise, may have some shared pleiotropic genetic basis. Therefore selection on this ‘multivariate set’ of traits may not be free to maximize all traits simultaneously, or reflect the patterns predicted under a fitness maximizing principle alone.Slide1

To understand the genetic architecture and evolution of social traits involved in conflict and cooperation we use a uniquely powerful and novel approach in which complimentary methods are integrated.

(1) Quantitative genetics: Because success in social interactions involves the concerted interaction of numerous traits expressed by the interacting individuals (including effects on and responses to the environment), their analysis is particularly well suited to quantitative genetics (QG) modelling approaches that explicitly incorporate the genetic relationships between traits when modelling phenotypic evolution.

(2) Experimental evolution and molecular genetics: Although QG models can generate rigorous predictions about the evolution and maintenance of social strategies, they are necessarily abstractions of the molecular genetic mechanisms underlying these strategies. To address this, we use an experimental evolution approach that combines controlled mutagenesis and selection for success in social interactions to examine how populations explore multivariate genetic space in response to the joint action of mutation and selection. This allows us to directly examine the process of social evolution and establish the link between molecular genetic variation and quantitative genetic variation. We will then begin to dissect and characterise the molecular and biochemical interactions among multigenic networks by identifying and characterising the genes underlying variation in social traits.

(3) Genome-wide identification of loci responsible for patterns of natural variation in social and non-social traits: We integrate high throughput phenotyping and large-scale genotyping of natural isolates to identify sequence variation associated with natural variation in fitness related traits. These data provide unprecedented insights into the genetic architecture of natural variation in social and other fitness related traits.

(4) Modelling: We integrate and develop theoretical models that will aid our understanding of the results, link each of the experimental components, and motivate experimentation through testable hypotheses.

We believe that these different approaches are highly complementary and that their integration is required to fully understand the multivariate elements of social evolution. We therefore propose to employ these techniques to understand the genetics of social evolution, using the social amoeba Dictyostelium discoideum as a model system.