Skip to main content

Quantitative Evolution

Final Report Summary - QUANTEVOL (Quantitative Evolution)

In the context of unprecedented anthropogenic forcing in the living world, the speed and mechanisms of evolution remain the key to many questions ranging from pathogens adaptation to the dynamics of biodiversity. To what extent can evolutionary theory make predictions? What would be the time horizon of such predictions? These are the general questions addressed with this project. Of course, evolutionary theory does not start from scratch and it has been a quantitative discipline for a long time. However, several key issues limit our ability to use evolutionary theory to make predictions. The first is that we lack a quantitative theory for predicting the effect of new mutations under different circumstances. The second is that we do not usually incorporate the diversity of mutational effects in models of adaptation. The third is that we rarely confront quantitatively this kind of prediction on patterns of long term adaptation in the laboratory and in nature. The aim of this project was to address these different issues.
We have only a limited idea on how predicting mutational input. Yet, mutations are ‘the raw material’ for evolution. The first achievement of this project was to extend a stand-alone and general approach to predict the distribution of mutational effects. This theory is a ‘top-down’ approach based on a n-dimensional fitness landscape model. We used it to statistically predict the effect of single mutants and the effect of the interaction of mutations at the same or different loci (dominance and epistasis, respectively). This theory was also used to predict the effect of mutations across environments and extended to incorporate mutational heterogeneity across loci, which delineates the conditions under which the genetic basis of adaptation can be predicted.
The second important achievement of this project was to confront this theory to the available data. Much more data are required to obtain a full picture, but so far the theory was impressively accurate to model weak to moderate mutational effects. New experiments were also set up in this project to specifically test the theory and better understand the variation of mutational effects across environments.
The third important achievement of the project was to elaborate new methods to measure fitness with better precision. Typically, precision in fitness measures is very rarely below 1%, even in demanding experimental assays. Such differences in fitness are however large from an evolutionary perspective and introduce considerable uncertainty in the prediction that can be made or tested. We developed a new method based on fluorescent bacteria and state-of-the-art statistical analysis, which allowed us to measure selection coefficients down to a precision of 2 x 10-4, a 50-fold improvement over standard methods and a 10-fold improvement over any previous study. We also developed a new method for measuring fitness from life history trait data. This method, called ‘Lifelihood’ is a generalized multi-event survival model, that will be extremely useful in a variety of context in evolution and ecology.
The fourth important achievement of the project was to investigate situation involving adaptation to abrupt environmental change. This question is directly relevant to the question of niche evolution in the context of global change. We investigated long term adaptation to acid conditions in bacteria (E. coli) in the laboratory. This experiment revealed that niche evolution can be accurately described using global deformation models (e.g. niche shifts, niche width variation), but that conventional scenarios (sliding niche model) need to be revised and account for plasticity evolution. We also investigated adaptation to increased temperature in the field using brine shrimps (Artemia) transplanted in tropical salterns as a model. This study indicated, that even on a significant time step (hundred of generations), little adaptation is occurring in the field, in contrast to laboratory situations. This is likely the consequence of the numerous trade-off that populations face in natural conditions compared to the laboratory, especially due to biotic interactions that we investigated and that had strong population impact even in the biodiversity-poor hypersaline artemia habitat.
Overall the project led to an integrative set of contributions, that together, made a significant step to turn evolutionary biology into a predictive rather than retrospective science.