The effect of genome organisation on selection efficiency in two contrasted plant species

The level of complexity and the extent of pleiotropy have important implications for our understanding of molecular evolutionary patterns across the genome. For example, under the Fisher Geometric Model, which assumes universal pleiotropy, all new mutations will have the same effect on fitness, and complexity governs the strength of selection. However, the extent of pleiotropy in biological systems is debated. In this study, we addressed this question by investigating how genome biology affects molecular evolution using whole genome data from Arabidopsis thaliana and Capsella grandiflora , two closely related Brassica species that vary significantly in their demography and mating system. Results were similar across both species. We found that several genome biology traits, including gene expression level, a trait related to pleiotropy, and gene network connectivity, a trait related to complexity, were predictive of the efficiency of selection and the mean strength and variance of selection acting against new deleterious variants, and that these patterns were not solely due to variation in local effective population size across the genome. We also found evidence that similar sites experience purifying selection at both long and short evolutionary timescales. Our results suggest that, across the genome, molecular evolutionary patterns are in agreement with the predictions of the FGM, albeit relaxing the assumption of universal pleiotropy.