Updating the resolution for 16S rRNA OTUs clustering reveals the cryptic cyanobacterial genus and species

Abstract

In processing metabarcoding data from the 16S rRNA amplicon sequencing, OTUs were commonly binned at 97% similarity, the criteria was once proposed as the limit for species delimitation for prokaryotes. This numeric guide is rather a tunable parameter and has recently been revised and increased up to 99% threshold facilitated with more 16S rRNA sequences available. In the study, we updated 97% identity to 99% in combination with monophyletic species concept rather than OTUs to evaluate cyanobacterial diversity from 11 freshwaters using metabarcoding technique. The cyanobacterial OTUs were first determined by aligning to the SILVA database, and further reassigned through the evolutionary placement algorithm (EPA). Inferred from the EPA, metabarcoding found diverse cyanobacterial community with at least 70 monophyletic species (113 OTUs) from 18 coccoid, 14 simple filamentous, 10 heterocytous genera, and Prochlorothrix by clustering OTUs at 97% similarity. When updating to 99% similarity, 5 cryptic genera were detected with 3 of them should be revised as new genera for they achieved monophyly in the phylogeny. In addition, updating to 99% similarity also detected 48 cryptic monophyletic species from 9 genera. In conclusion, updating 97% identity to 99% in combination with monophyletic species found cryptic genera and species, and provided a much clearer separation of cyanobacterial community between lakes and reservoirs and also between seasons. The combined methods improved the use of metabarcoding technique to elucidate field cyanobacterial community.