Mathematical models for the evolution of species in the era of genetic data
Abstract: One of the aims of population genetics is to date and quantify major events that arrived during the evolution of species. The DNA sequencing techniques and the increasing amount of genomic data currently available are expanding the horizons of the field. Over the last decade, a wide range of methods allowing to infer past population size changes from genome-wide data have been developed in order to better exploit this new data. In this talk, I will discuss some mathematical models at the bottom of these inference methods. I will start by considering a population evolving under the Wright-Fisher model. Then I will introduce the Coalescent which is a stochastic process allowing to trace back the ancestry of the individuals. The diversity observed in genomic data can be modeled as the result of a Poisson process when mutations are added to the coalescence tree. At the same time, the topology of the coalescence tree is very sensitive to population size changes that occurred in the past. This idea lays the foundations of many inferential frameworks used to reconstruct the demographic history from genetic data. I will illustrate two inference methods based on the above reasoning. Finally, I will comment the results obtained when we apply one inferential framework to human data.
Coffee will be served from 10h00 – 10h30.