Confident phylogenetic identification of uncultured prokaryotes through long read amplicon sequencing of the 16S‐ITS‐23S rRNA operon

Amplicon sequencing of the 16S rRNA gene is the predominant method to quantify microbial compositions and to discover novel lineages. However, traditional short amplicons often do not contain enough information to confidently resolve their phylogeny. Here we present a cost‐effective protocol that amplifies a large part of the rRNA operon and sequences the amplicons with PacBio technology. We tested our method on a mock community and developed a read‐curation pipeline that reduces the overall read error rate to 0.18%. Applying our method on four environmental samples, we captured near full‐length rRNA operon amplicons from a large diversity of prokaryotes. The method operated at moderately high‐throughput (22286–37,850 raw ccs reads) and generated a large amount of putative novel archaeal 23S rRNA gene sequences compared to the archaeal SILVA database. These long amplicons allowed for higher resolution during taxonomic classification by means of long (∼1000 bp) 16S rRNA gene fragments and for substantially more confident phylogenies by means of combined near full‐length 16S and 23S rRNA gene sequences, compared to shorter traditional amplicons (250 bp of the 16S rRNA gene). We recommend our method to those who wish to cost‐effectively and confidently estimate the phylogenetic diversity of prokaryotes in environmental samples at high throughput.