Abstract
The study of variations in the recombination rates along genomes has revealed the existence of hotspots in many organisms. In most mammals, these hotspots are evolutionarily labile and are regulated by the PRDM9 protein. Although the Prdm9 gene appears to have been present in the common ancestor of animals, the evolution of its function in regulating hotspots remains unknown. The functional copy of PRDM9 described in mammals appears to have been repeatedly lost during animal evolution, as in birds, where hotspots are evolutionarily stable. The relationship between the presence/absence of PRDM9 and recombination landscapes remains unclear at the animal level, as efforts to date have mainly focused on mammals. The aim of my thesis is to investigate the role of PRDM9 in the recombination landscapes of non-mammalian animals that possess the PRDM9 protein, in particular salmonids and neopteran insects, using linkage disequilibrium-based methods. I was able to show that the LDhelmet method is reliable for inferring recombination rates when genomic data are of good quality and polymorphism is high, but its performance declines rapidly when the effective population size is low or the population is not in demographic equilibrium. Recombination events are regulated by the PRDM9 protein in salmonids, suggesting that the mammalian function of PRDM9 is ancestral to vertebrates. However, it appears that the default system (i.e. without the PRDM9 protein) is also involved in determining some of the salmonid hotspots, with the two systems not being mutually exclusive. The recombination landscapes in insects are different from those described in vertebrates and appear to depend on the characteristics of the methylation landscapes, which vary greatly between taxonomic groups. Finally, the presence of PRDM9 in phasmids and water bugs seems to be associated with unstable hotspots, as in vertebrates. These results have allowed me to gain a better understanding of the determinants and evolutionary dynamics of recombination landscapes in animals, and this work should be continued in invertebrates.