In this study we found that population structure, independently from the degree of connectivity between demes and the migration matrix relating them, intrinsically determines a variation in the rate of coalescence events through time. We showed that the intensity and the direction(s) of such variation related to the demographic parameters of the meta-population in a predictable way. Our results highlight the importance of detecting population structure (which depends on LHT among other factors) before performing any demographic inferences but, at the same time, they reveal the utility of unstructured models to describe the shape of the gene genealogy, which is the final product of the evolutionary history of a species. A combination of structured and unstructured models (better if non-parametric) is therefore the key to best characterize the evolutionary history of a species. We call for a change in perspective when investigating the demographic history of a species: the focus should be put in the reconstruction of the variation of both N and m through time, which requires certainly new methodological development and probably more data.
Acknowledgement
We are grateful to the Genotoul bioinformatics platform Toulouse
Midi-Pyrenees (Bioinfo Genotoul; http://bioinfo.genotoul.fr/) for
providing computing resources. We are indebted to Oscar Lao for fruitful
discussions and careful reading of the manuscript. This work was
supported by two ATM grants (2016 and 2017) from the Muséum National
d’Histoire Naturelle to S.M.