High-accuracy meets high-throughput for microbiome profiling with near full-length 16S rRNA amplicon sequencing on the Nanopore platform

Abstract Amplicon sequencing of small subunit (SSU) rRNA genes is a foundational method for studying microbial communities within various environmental, human, and engineered ecosystems. Currently, short-read platforms are commonly employed for high-throughput applications of SSU rRNA amplicon sequencing, but at the cost of poor taxonomic classification. The low-cost Oxford Nanopore Technologies (ONT) platform is capable of sequencing full-length SSU rRNA genes, but the lower raw-read accuracies of previous ONT sequencing chemistries have limited accurate taxonomic classification and de novo generation of operational taxonomic units (OTUs) and amplicon sequence variants (ASVs). Here, we examine the potential for Nanopore sequencing with newer (R10.4+) chemistry to provide high-throughput and high-accuracy full-length 16S rRNA gene amplicon sequencing. We present a sequencing workflow utilizing unique molecular identifiers (UMIs) for error-correction of SSU rRNA (e.g. 16S rRNA) gene amplicons, termed ssUMI. Using two synthetic microbial community standards, the ssUMI workflow generated consensus sequences with 99.99% mean accuracy using a minimum UMI subread coverage threshold of 3x, and was capable of generating error-free ASVs and 97% OTUs with no false-positives. Non-corrected Nanopore reads generated error-free 97% OTUs but with reduced detection sensitivity, and also generated false-positive ASVs. We showcase the cost-competitive and high-throughput scalability of the ssUMI workflow by sequencing 90 time-series samples from seven different wastewater matrices, generating ASVs that were tightly clustered based on sample matrix type. This work demonstrates that highly accurate full-length 16S rRNA gene amplicon sequencing on Nanopore is possible, paving the way to more accessible microbiome science.