Statistical questions in experimental evolution

Recent advances in the mathematical analysis of models describing evolutionary dynamics are rapidly increasing our ability to make precise quantitative predictions. These advances have created a growing need for corresponding improvements in our ability to observe evolutionary dynamics in laboratory evolution experiments. High-throughput experimental methods are particularly crucial, in order to maintain many replicate populations and measure statistical differences in evolutionary outcomes at both phenotypic and genomic levels. In this paper, I describe recent technical developments which have greatly increased the throughput of laboratory evolution experiments, and outline a few promising directions for further improvements. I then highlight a few ways in which these new experimental methods can help to answer simple statistical questions about evolutionary dynamics, and potentially guide future theoretical work.