Making sense of the virome in light of evolution and ecology

Abstract

Understanding the patterns and drivers of viral prevalence and abundance is of key importance for understanding pathogen emergence. Over the last decade, metagenomic sequencing has exponentially expanded our knowledge of the diversity and evolution of viruses associated with all domains of life. However, as most of these 'virome' studies are primarily descriptive, our understanding of the predictors of virus prevalence, abundance and diversity, and their variation in space and time, remains limited. For example, we do not yet understand the relative importance of ecological predictors (e.g. seasonality and habitat) versus evolutionary predictors (e.g. host and virus phylogenies) in driving virus prevalence and diversity. Few studies are set up to reveal the factors that predict the virome composition of individual hosts, populations or species. In addition, most studies of virus ecology represent a snapshot of single species viromes at a single point in time and space. Fortunately, recent studies have begun to use metagenomic data to directly test hypotheses about the evolutionary and ecological factors which drive virus prevalence, sharing and diversity. By synthesizing evidence across studies, we present some over-arching ecological and evolutionary patterns in virome composition, and illustrate the need for additional work to quantify the drivers of virus prevalence and diversity.

Keywords: virome; virus communities; virus ecology; virus evolution; virus metagenomics; viruses.

Figure 1.

Host and virus species level and phylogenetic effects on virus prevalence and viral host range. The y-axis represents a hypothetical virus phylogeny and the x-axis a hypothetical host phylogeny. Asterisk indicates model interaction terms. Each panel represents different possible scenarios. (A) The incidence and prevalence of viruses across host species is predictable by host phylogeny (i.e. closely related host species have a similar incidence of viruses). (B) The incidence and prevalence of viruses across host species is predictable by virus phylogeny (i.e. closely related viruses have a similar infectivity across host species). (C) Certain host species are inherently more or less susceptible to viruses, in a way not predictable by the host phylogeny (i.e. due to ecological or physiological traits). (D) Certain viruses are particularly infectious, or not, irrespective of host species, in a way not predictable by virus phylogeny. (E) Related hosts have similar incidences of clades of related viruses (i.e. virus incidence and prevalence is predictable by both host and virus phylogenies). (F) Related hosts have similar incidences of some viruses, but not all, and not in a phylogenetically predictable manner. (G) Related viruses show similar infectivity to only some host species—not all—and not in a way predicted by host phylogeny. (H) Host susceptibility depends on specific host × virus interactions not predictable by either host or virus phylogeny. Based on [34,53].

Figure 2.

The ecological and evolutionary drivers of viromes. Viromes can be considered at multiple levels: individual organisms, populations, species or whole ecosystems. Factors influencing the virome may interact. For example, seasonal changes in host range may coincide with seasonal peaks in infection burden, with coinfection interactions shifting components of the virome.