Rapid phylogenetic analysis using open reading frame content patterns acquired by Oxford nanopore sequencing.

Abstract

Phylogenetic analysis based on core genome single nucleotide polymorphisms (cgSNPs) using whole-genome sequencing (WGS) is increasingly used in epidemiological investigations of bacteria. The approach, however, is both resource intensive and time-consuming. Oxford Nanopore Technologies (ONT) sequencing is capable of real-time data analysis but the high error rate hampers its application in cgSNP-based phylogenetic analysis. Here, we developed a cgSNP-independent phylogenetic analysis method using ONT read assemblies by focusing on open reading frame (ORF) content patterns. WGS data of 66 Enterobacter hormaechei strains acquired by both ONT and Illumina sequencing and 162 strains obtained from NCBI database were converted to binary sequences based on the presence or absence of ORFs using BLASTn. Phylogenetic trees calculated from binary sequences (ORF trees) were compared with cgSNP trees derived from Illumina sequences. Clusters of closely related strains in the cgSNP trees formed comparable clusters in the ORF trees built with binary sequences, and the tree topologies between them were similar based on Fowlkes-Mallows index. The ORF-based phylogenetic analysis using ONT sequencing may be useful in epidemiological investigations and offer advantages over the cgSNP-based approach. Conversion of assembled WGS data to binary sequences based on the presence or absence of ORFs circumvents read error concerns with ONT sequencing. Since ONT sequencing generates data in real time and does not require major investment, this ORF-based phylogenetic analysis method has the potential to enable phylogenetic and epidemiological analysis at the point of care.