Ecology and Evolution Nature and Landscape Conservation
Tradeoffs in Method Selection for Analyzing Full‐Length 16S Data From Nanopore Sequencing in Defined and Environmental Microbiomes
Abstract
ABSTRACT Bacteria are found in nearly every environment on earth, and the ability to characterize bacterial communities is key to investigating a variety of scientific questions. One method for characterizing and comparing bacterial communities is to amplify and sequence the 16S ribosomal RNA gene and to use these sequences (A) to determine phylogenetic relationships within and between samples and (B) to assign a taxonomic identification to each sequence by comparison to reference databases. Nanopore sequencing technology has increased the accessibility and affordability of sequencing for full‐length 16S ribosomal RNA genes, but existing workflows for analyzing these data are either not developed for the unique error profiles of Nanopore data or are recently developed and incompletely characterized. A detailed comparison between Nanopore‐specific metabarcoding pipelines is needed to illustrate how the tradeoffs inherent in all metabarcoding workflows manifest in these pipelines, and to inform workflow selection for specific applications. Here we compare three pipelines: Emu, PRONAME, and wf‐metagenomics to analyze microbiomes of simple and complex bacterial communities. We find that all three pipelines provide similar conclusions for environmental samples with respect to alpha and beta diversity and order‐level taxonomy while reporting different taxonomic classifications and richness at the family, genus, and species levels. We find that all three can reliably identify genera in simple, defined communities. The comparisons provided here will aid those using Nanopore Sequencing for 16S metabarcoding to identify an appropriate analysis workflow for their intended question or application.