Special seminar - Bram van Dijk: The ecology and evolution of Virtual Microbes

Abstract:

Even on the scale of single grains of soil, microbial ecosystems are incredibly diverse. In my talk I will focus on the factors that help or hinder ecological interactions from evolving. Although we have an increasing understanding of the components that make up a microbial ecosystem (cells, genes, metabolites, etc.), we have not yet developed our intuition on what this system looks like when the wheels are actually turning. We have developed a multi-level simulation framework called "Virtual Microbes" with which we can shine some light on these dynamics. I will present our work where we have mimicked Richard Lenski's long-term evolution experiment in silico. We find that different "pre-evolved" Virtual Microbe wild types (WTs) adapt to anticipate the regularity of the serial transfer protocol by accurately fitting their growth dynamics to the 24 cycle. This anticipation is achieved in a variety of mechanistically different ways, one of which is the formation of a microbial collective, where two strains have adapted on their own temporal niche. While some WTs repeatedly evolve the same solution to anticipate the serial transfer protocol, others find a broad range of different solutions. In a follow-up study, we have found that these historical contingencies are also present when evolving Virtual Microbes on a single resource without a daily resource cycle, and that small differences in the evolutionary history of microbes can blow up to shape the outcome of the entire ecosystem. In short, historical contingencies in the predictability of evolution, and what kind of ecology evolves, is itself a historical contingency.