De novo species identification using 16S rRNA gene nanopore sequencing
Angell, Inga Leena; Nilsen, Morten; Carlsen, Karin C. Lødrup; Carlsen, Kai-Håkon; Hedlin, Gunilla; Jonassen, Christine M; Marsland, Benjamin; Nordlund, Bjørn; Rehbinder, Eva Maria; Schinagl, Carina Madelen; Skjerven, Håvard Ove; Staff, Anne Cathrine; Söderhall, Cilla; Vettukattil, Muhammad Riyas; Rudi, Knut
Peer reviewed, Journal article
MetadataVis full innførsel
OriginalversjonPeerJ. 2020, . 10.7717/peerj.10029
Nanopore sequencing is rapidly becoming more popular for use in various microbiotabased applications. Major limitations of current approaches are that they do not enable de novo species identification and that they cannot be used to verify species assignments. This severely limits applicability of the nanopore sequencing technology in taxonomic applications. Here, we demonstrate the possibility of de novo species identification and verification using hexamer frequencies in combination with kmeans clustering for nanopore sequencing data. The approach was tested on the human infant gut microbiota of 3-month-old infants. Using the hexamer k-means approach we identified two new low abundant species associated with vaginal delivery. In addition, we confirmed both the vaginal delivery association for two previously identified species and the overall high levels of bifidobacteria. Taxonomic assignments were further verified by mock community analyses. Therefore, we believe our de novo species identification approach will have widespread application in analyzing microbial communities in the future.