What Microbial Population Genomics Has Taught Us About Speciation

Abstract

Population genomics has emerged as a valuable tool to define and delimit species and to understand the mechanisms that drive and maintain speciation. Species and speciation have been notoriously difficult to study in microbes owing to their asexual reproduction, promiscuous horizontal gene transfer, and obscure microscopic niches. Over the past few years, whole-genome sequencing of closely related, locally co-occurring populations of microbes, combined with simulations and modelling, has revealed certain general features of microbial speciation: it is usually driven by divergent natural selection between distinct ecological niches (a form of the ecological species concept), and species distinctness is maintained by barriers to gene flow (a form of the biological species concept). In some cases, gene-flow barriers may come about as a natural consequence of ecological specialization. Although these features appear to be quite general, there are exceptions. Trivially, barriers to gene flow cannot be used to delimit clonal populations where there is negligible gene flow. More interestingly, it is unclear whether other barriers to gene flow, such as genetic incompatibilities or differences in phage-host range, are able to drive speciation in the absence of other selective pressures. Here, I discuss the extent to which speciation is driven by natural selection, gene-flow barriers, or a combination of the two, drawing on recent examples from bacterial and archaeal population genomics, experimental evolution, and modelling. I then describe how population genomic data can be used to define and delimit species boundaries, based upon nucleotide identity cutoffs or upon discontinuities in gene flow. Despite important limitations and caveats, delimitation methods provide a useful starting point for more detailed investigation into the genetic and ecological basis of speciation.