Fast genome-based species delimitation: Enterobacterales and beyond

Average Nucleotide Identity (ANI) is becoming a standard measure for bacterial species delimitation. However, its calculation can take orders of magnitude longer than fast similarity estimates based on sampling of short nucleotides, compiled into so-called sketches. These estimates are widely used and correlate well with ANI. However, they might not be as accurate. Thus, we compared two sketching programs, mash and dashing, against ANI, in delimiting species among publicly available Esterobacterales genomes. Receiver Operating Characteristic (ROC) curve analysis found all three programs to be highly accurate, with Area Under the Curve (AUC) values of 0.99, indicating almost perfect species discrimination. Subsampling to reduce over-represented species, reduced these AUC values to 0.92. Focused tests with ten genera represented by more than three species, also showed almost identical results for all methods. Shigella showed the lowest AUC values (0.68), followed by Citrobacter (0.80). All other genera, Dickeya, Enterobacter, Escherichia, Klebsiella, Pectobacterium, Proteus, Providencia and Yersinia, produced AUC values above 0.90. The species delimitation thresholds varied, with species distance ranges in a few genera overlapping the genus ranges of other genera. Mash was able to separate the E. coli + Shigella complex into 25 apparent phylogroups. Testing mash for species separation in genera outside Enterobacterales showed AUCs above 0.95, again with different thresholds for species delimitation within each genus. Overall, our results suggest that fast estimates of genome similarity are as good as ANI for species delimitation. Therefore, these fast estimates might suffice for determining the role of genomic similarity in bacterial taxonomy.