Comparative analysis of deep mutational scanning datasets in enteroviruses A and B identifies functional divergence and therapeutic targets.

Deep mutational scanning (DMS) can define functional constraints acting on viral proteomes by quantifying the effects of mutations on viral fitness. However, DMS analyses do not discern type-specific from species-level constraints, limiting their utility in understanding how selective pressures change as viral families diversify. Here we show that comparison of DMS datasets from related viruses can overcome these limitations. By contrasting two proteome-wide DMS datasets from prototypical members of the enterovirus A and B species, we identify evolutionary constraints at the species level to occur across core enzymatic machinery and capsid assembly interfaces. In contrast, type-level constraints are observed across host-interaction sites in both structural and non-structural proteins. Furthermore, we find DMS data to reflect both type- and species-level evolutionary signatures in nature yet diverge at conserved hotspots subjected to selection pressures that are lacking in vitro. Finally, we highlight the utility of comparative DMS studies for drug discovery by identifying a mutationally constrained pocket in the 2C helicase that is conserved across all major human enterovirus species. Our findings provide a framework for dissecting evolutionary pressures acting at different evolutionary scales and for guiding the rational design of broad-spectrum therapeutics with high barriers to resistance.