There is no evidence of a universal genetic boundary among microbial species

Abstract

A fundamental question in studying microbial diversity is whether there is a species boundary and if the boundary can be delineated by a universal genetic discontinuity. To address this question, Jain et al. computed the pairwise average nucleotide identity (ANI) of 91,761 microbial (bacterial and archaeal) genomes (the 90K genome dataset) and found that the ANI values from the 8 billion comparisons follow a strong bimodal distribution1. The authors concluded that a clear genetic discontinuum and species boundary were evident from the unprecedented large-scale ANI analysis. As a result, the researchers advocated that an ANI of 95% can be used to accurately demarcate all currently named microbial species. While the FastANI program described in the paper is useful, we argue that the papers conclusion of a universal genetic boundary is questionable and resulted from the substantial biased sampling in genome sequencing. We also caution against being overly confident in using 95% ANI for microbial species delineation as the high benchmarks reported in the paper were inflated by using highly redundant genomes.