BioProject

Nanopore technology is a promising platform for the introduction of metagenomic amplicon sequencing into the clinical microbiology diagnostics. More...

Nanopore technology is a promising platform for the introduction of metagenomic amplicon sequencing into the clinical microbiology diagnostics. However, few benchmarks have been performed to evaluate the accuracy of databases used for taxonomic classification of the sequencing reads. Using two different commercial software tools (Qiagen CLC Genomics Workbench and Nanopore EPI2ME), we compared five different databases for the taxonomic identification of 16S and 23S bacterial amplicons (QMI, LSU, SSU, 16S and WIMP), and seven different databases for classification of 18S fungi reads (UNITE, UNITE DYNAMIC, NCBI 18S, NCBI ITS, NCBI 28S, LSU, and WIMP). We observed large discrepancies in the accuracy of the database for the identification of samples of known composition (community standards or pure cultures) when used with Nanopore sequence results. Results ranged from 100 % correct identification at the genus level for the QMI database to 15% for the SSU database for bacterial pure cultures. For mock community samples, the QMI database also performed very well with only 1% misidentified reads, whereas almost 30% of the reads are misidentified by the EPI2Me default database (WIMP). These large differences between the databases highlight the importance of properly validating bioinformatics pipelines as well as databases used for clinical microbiology diagnostic usage. Additional benchmarking tools and studies are required to identify and correct missing elements in existing databases. Less...