Microbiomes change the rules of host evolution
Microbiomes are pervasive determinants of host phenotype and fitness, yet their consequences for host population genetics remain poorly understood. Here we show that microbiome-associated population structure can fundamentally alter the evolutionary dynamics of host populations. We develop a general framework that maps hosts distributed among an arbitrary number of microbial states onto an effective population size, $N_{\mathrm{eff}}$, and effective selection coefficient, $s_{\mathrm{eff}}$, thereby reducing potentially high-dimensional host--microbiome dynamics to the quantities governing genetic drift and selection. Applying the framework to five empirically anchored host--microbe systems spanning ciliates, insects, and plants with bacterial, algal, and fungal associates reveals substantial departures from classical population-genetic predictions. Microbiome-associated structure can considerably alter both the effective strength of genetic drift and the strength of selection acting on host genetic variation, shifting populations into evolutionary regimes that differ markedly from those predicted from census size and intrinsic mutational effects alone. These changes can have profound consequences for evolutionary outcomes, including either suppression or enhancement of selection for both advantageous and disadvantageous mutations, sometimes reaching orders-of-magnitude differences. The effects are not determined simply by whether a microbial association is beneficial or detrimental, but by the ecological processes governing microbial acquisition, transmission, loss, and host demography. Our results provide a population-genetic extension of the holobiont perspective: when microbial associations generate distinct demographic states among hosts, they can alter the evolutionary forces acting on the host genome without requiring the host and microbiome to constitute a single hereditary unit.