The interplay between migration and selection on the dynamics of pathogen variants

The fitness advantage of an emerging pathogen variant is typically estimated from its change in frequency over time. This approach relies on the assumption that the pathogen spreads in a well-mixed population, where frequency changes are solely driven by selection. Yet, spatial structure can have major consequences on the spread of new variants. Here we model the change in frequency across time and across space of a new pathogen variant spreading in a two-patch host metapopulation. Crucially, we show that even small rates of migration can interfere with selection and may bias the estimation of fitness. We illustrate this effect with the evolution of SARS-CoV-2 by contrasting the spread of the Alpha and the Delta variants in England. We contend that the observed heterogeneity of fitness estimates across space could result from the influence of pathogen migration. This work highlights the need of a comprehensive theoretical framework accounting for the interplay between selection and migration on the spread of new pathogen variants.