Comparison of Illumina and Oxford Nanopore Technology systems for the genomic characterization of Streptococcus pneumoniae

Abstract

Whole-genome sequencing (WGS) is an invaluable tool that enables high-resolution genotyping to precisely identify bacterial strains. It is particularly significant for highly pathogenic bacteria such as Streptococcus pneumoniae, a worldwide leading cause of mortality and morbidity. Illumina sequencing is highly established for S. pneumoniae, while Oxford Nanopore Technologies (ONT) data are limited. Hence, evaluating ONT-only data is needed. We aimed to compare the Illumina and ONT systems for S. pneumoniae sequencing. Moreover, we aimed to explore whether the newer chemistry from ONT with R10.4.1 flow cells improves the data outputs from long-read sequencing. S. pneumoniae bacteria were isolated from hospitalized patients with invasive pneumococcal disease (IPD) and serotyped by multiplex PCR. Resistance profiles were determined with anti-microbial susceptibility testing. A total of 27 isolates were sequenced using ONT Mk1c with R9.4.1 flow cells and Kit10 chemistry (ONT_V10) and the Illumina Miseq system. Illumina and ONT data were compared, and hybrid assembly was assessed. ONT sequencing was additionally performed with R10.4.1 flow cells and Kit14 chemistry (ONT_V14) in 12 isolates. S. pneumoniae identification, serotyping, AMR, and GPSC prediction were successfully achieved using ONT sequencing. The ONT_V14 chemistry significantly improved both MLST and pbp prediction in long-read sequencing. Overall, the hybrid assembly produced circular and contiguous genomes with high N50 parameters. Moreover, long-read assembly followed by short-read polishing is a fast and reliable approach for hybrid assembly at ONT sequencing depth >100×. For ONT sequencing depth <50×, tools that perform short-read-first assembly, such as Unicycler are recommended.IMPORTANCEThis study provides a detailed evaluation of whole-genome sequencing technologies and bioinformatics pipelines for the characterization of Streptococcus pneumoniae. It represents an in-depth investigation of Illumina and Oxford Nanopore technologies (ONT) systems for bacterial sequencing. It sheds light on the performance of each platform in various aspects of sequencing, including raw and assembly statistics, capsular typing, pbp typing, GPSC, AMR, and MLST prediction. This study offers a comprehensive overview of S. pneumoniae genomics and a guide for clinical and research laboratories seeking to adopt bacterial sequencing by providing important considerations when choosing sequencing platforms and analysis pipelines. We report a strong case for the implementation of WGS in the clinical setting, based on its high concordance with conventional molecular and phenotypic methods. Furthermore, the flexibility and portability of the investigated pipelines facilitate their use in clinical applications.

Keywords: Illumina; ONT; Streptococcus pneumoniae; genomics; hybrid assembly; whole-genome sequencing.

Fig 1

Quality of raw data and assembled genomes obtained from Streptococcus pneumoniae isolates (n = 27) sequenced by both Illumina and ONT_V10 platforms. Raw data were compared between both techniques according to the median sequencing depth (A), number of reads (B), and the averages of read length (C) and quality scores (D). Panel E represents the median number of contigs in assemblies generated from Illumina or ONT data. Panel F represents the number of contigs obtained from ONT sequencing according to sequencing depth cut-offs of 50× and 100×. Panel G plots the variation in contiguity according to ONT sequencing depth. Significant values are indicated with **P < 0.01 or ***P < 0.001 (Mann-Whitney test). Abbreviations: bp, Base pairs; ONT_V10, Oxford Nanopore Technologies (R9.4.1 flow cells/ Kit10 chemistry).

Fig 2

Illustrated overview of the workflow utilized to compare hybrid assemblies generated by Unicycler and Pilon for S. pneumoniae isolates (n = 27). S. pneumoniae bacteria were sequenced using the ONT Mk1c system with R9.4.1 flowcells and Kit10 chemistry (1) and the Illumina Miseq system (2). Illumina and ONT_V10 FastQ files were processed using Unicycler, which utilizes long reads to scaffold a short-read assembly graph to completion (3). ONT_V10 FASTA files were indexed and used to align Illumina reads (4). The resulting SAM files were processed by samtools and utilized by Pilon for hybrid assembly. The genome quality results from both hybrid assembly approaches were compared (5). Abbreviations: ONT_V10, Oxford Nanopore Technologies (R9.4.1 flow cells/ Kit10 chemistry).

Fig 3

Maximum-likelihood phylogenetic tree of Streptococcus pneumoniae isolates (n = 27) constructed from genome data generated by hybrid assembly of ONT_V10 and Illumina data. The tree nodes are colored by year of collection. Abbreviations: MDR, multidrug resistance; PILI1, type 1 pili; PILI2, type 2 pili.