View Program - 2022 Population, Evolutionary, and Quantitative Genetics Conference

Measuring fitness-associated genetic change in a natural population is a persistent goal for evolutionary biologists. Quantitative genetics (QG) provides theory and methods to extract information about the raw material on which natural selection can work, i.e., the additive genetic variance. So far, Milot et al. (2011) remain the only study in humans to robustly show contemporary evolution in response to selection, using the preindustrial île aux Coudres (IAC) population. It was determined by QG analysis combined with pedigree information that the age at first reproduction (AFR) of married women, which evolved from 26 y to 22 y within 5 to 8 generations, was partially due to a genetic change in response to selection. We hypothesized that fluctuations in environmental conditions might induce changes in the genetic architecture of AFR and hence its rate of evolutionary change, mainly through genotype-by-environment interactions or GxE. We tested this hypothesis using the IAC population by incorporating the infant mortality rate as a proxy of the harshness of conditions in early life. Our results detected the presence of GxE underlying the variation in AFR and relative fitness in this population. Based on our predictions using the Roberston-Price covariance, GxE interactions led to an increase in the expected per-generation genetic change in AFR under harsh early-life conditions compared to more benign conditions. Deep-rooted genealogical information is available for the historical French-Canadian population, and it is vital to determine the reliability of this dataset in terms of QG parameter estimation. The performed power and precision analysis on the reconstructed pedigrees revealed that most of the datasets were powerful enough to detect a simulated genetic effect. Precision and accuracy suffered greatly when the model did not include all sources of similarity between relatives (e.g., similarity due to familial environment) regardless of the sample size and depth of the pedigree. Finally, we tested if a response to selection in AFR also occurred in another larger preindustrial French-Canadian population, the Charlevoix population. Predictions point to a weak expected genetic response to natural selection in AFR in this second population. The fact that the detected temporal change in the average breeding values of AFR is equally as likely under a scenario of drift alone constitutes a robust demonstration of evolutionary stasis.

Evolutionary change in age at first reproduction in a preindustrial human population is faster at times of high infant mortality