Patterns of genetic diversity in socially structured populations : an individual-based approach

Tese de doutoramento, Biologia (Biologia Populacional), Universidade de Lisboa, Faculdade de Ciências, 2016Natural populations consist of spatial aggregations of interacting individuals shaped by geographic, ecological or behavioral factors. Relatively simple models have been proposed in population genetics to study how this subdivision affects the genetic diversity and persistence of populations in discontinued landscapes. Classical population genetic models view populations as networks of discrete entities (wherein mating occurs at random – demes) connected by migration. These models have demonstrated that population subdivision has implications on the effective population size, mean coalescence time and genetic variation. Although, in many species, demes are further subdivided into age-structured groups of individuals with complex mating strategies, social-structure has received little attention. This thesis focus on the genetic consequences of social structure. In particular, we asked whether social structure leads to genetic patterns that differ from those predicted by classical models, and to which extent ignoring social structure can bias inferences from real populations. An individual-based simulation framework was developed to investigate the effects of sex-biased dispersal and complex mating systems on genotypic frequencies, genetic diversity and gene genealogies. We found that social structure leads to an excess of heterozygotes within social groups (outbreeding) that is not detected when social groups are ignored (common practice in many empirical studies). Furthermore, we show that incorrect conclusions about inbreeding or random-mating may be drawn if social structure is not explicitly taken into account. This framework was applied to study a social species, the lemur Propithecus tattersalli. Simulations fitted the empirical results indicating that, in this species, social subdivision decreases inbreeding to a great extent. This allows this species to maintain high levels of individual diversity in its highly fragmented habitat and these results may be important for other endangered species. This study has also shown that social structure may bias inferences of past demographic events, often leading to spurious signals of expansions rather than bottlenecks. This work contributed to a better understanding of the effects of sociality, showing that social structure shapes the genetic diversity of populations in ways that cannot be predicted by classical genetic models.LABEX (Laboratoires d’Excellence), projeto TULIP (ANR-10-LABX-41); LIA BEEG-B (Laboratoire International Associé - Bioinformatics, Ecology, Evolution, Genomics and Behaviour; CNRS (Centre International de la Recherche Scientifique