Abstract
BACKGROUND: Amplicon sequencing on Illumina sequencing platforms leverages their deep sequencing and multiplexing capacity but is limited in genetic resolution due to short read lengths. While Oxford Nanopore or Pacific Biosciences sequencing platforms overcome this limitation, their application has been limited due to higher error rates or lower data output. RESULTS: In this study, we introduce an amplicon sequencing workflow, i.e., NanoAmpli-Seq, that builds on the intramolecular-ligated nanopore consensus sequencing (INC-Seq) approach and demonstrate its application for full-length 16S rRNA gene sequencing. NanoAmpli-Seq includes vital improvements to the INC-Seq protocol that reduces sample processing time while significantly improving sequence accuracy. The developed protocol includes chopSeq software for fragmentation and read orientation correction of INC-Seq consensus reads while nanoClust algorithm was designed for read partitioning-based de novo clustering and within cluster consensus calling to obtain accurate full-length 16S rRNA gene sequences. CONCLUSIONS: NanoAmpli-Seq accurately estimates the diversity of tested mock communities with average consensus sequence accuracy of 99.5% for 2D and 1D2 sequencing on the nanopore sequencing platform. Nearly all residual errors in NanoAmpli-Seq sequences originate from deletions in homopolymer regions, indicating that homopolymer aware base calling or error correction may allow for sequencing accuracy comparable to short-read sequencing platforms.