Understanding the evolution of all living organisms is one of the fundamental challenges in biology. The phylogenetic analysis of whole genomes has already proven very useful to decipher not only their history, but also their organization and function. Indeed, the accuracy of these inferences is intimately related to the quality of the models assumed. Despite important advances, models of genome evolution are still in its infancy, and more realistic models are needed to provide more precise and reliable inferences from genome data. In addition, a number of phylogenomic algorithms have been proposed to estimate phylogenies from complete genomes based on different genomic features. Although the application of these methods has already led to critical conclusions regarding the tree of life, the relative performance of these algorithms has not been properly evaluated yet. The first objective of this grant is to develop more realistic models of genome evolution, integrating changes in gene content and changes in gene sequences, and allowing for model variation along different branches of the phylogeny. In order to avoid model overparameterization, a statistical framework for the selection of best-fit models of genome evolution for the data at hand will also be implemented. Genome data simulated under these models will be used to compare the performance of different phylogenomic algorithms. Optimized phylogenomic strategies will then be applied to available genomes in order to decipher unresolved portions of the tree of life. Finally, all the bioinformatic tools developed under this grant will be made freely available to the scientific community.