Determining Streptococcus suis serotype from short-read whole-genome sequencing data

BMC Microbiology

Research Article

Daisuke Takamatsu¹  Sonia Lacouture²  Marcelo Gottschalk²  Taryn B. T. Athey³  Sarah Teatero³  Nahuel Fittipaldi⁴ 

[1] Bacterial and Parasitic Diseases Research Division, National Institute of Animal Health, National Agriculture and Food Research Organization, Tsukuba, Japan;The United Graduate School of Veterinary Sciences, Gifu University, Gifu, Japan;Groupe de Recherche sur les Maladies Infectieuses du Porc, Faculty of Veterinary Medicine, University of Montreal, St-Hyacinthe, QC, Canada;Public Health Ontario Toronto Laboratory, 661 University Avenue, Toronto, ON, M5G 1M1 Canada;Public Health Ontario Toronto Laboratory, 661 University Avenue, Toronto, ON, M5G 1M1 Canada;Department of Laboratory Medicine and Pathobiology, Faculty of Medicine, University of Toronto, Toronto, ON, Canada;

关键词: Streptococcus suis;    Serotyping;    Whole-genome sequencing;    Short-reads;   

DOI  :  10.1186/s12866-016-0782-8

 received in 2016-02-24, accepted in 2016-07-15,  发布年份 2016

来源: Springer

【 摘 要 】

BackgroundStreptococcus suis is divided into 29 serotypes based on a serological reaction against the capsular polysaccharide (CPS). Multiplex PCR tests targeting the cps locus are also used to determine S. suis serotypes, but they cannot differentiate between serotypes 1 and 14, and between serotypes 2 and 1/2. Here, we developed a pipeline permitting in silico serotype determination from whole-genome sequencing (WGS) short-read data that can readily identify all 29 S. suis serotypes.ResultsWe sequenced the genomes of 121 strains representing all 29 known S. suis serotypes. We next combined available software into an automated pipeline permitting in silico serotyping of strains by differential alignment of short-read sequencing data to a custom S. suis cps loci database. Strains of serotype pairs 1 and 14, and 2 and 1/2 could be differentiated by a missense mutation in the cpsK gene. We report a 99 % match between coagglutination- and pipeline-determined serotypes for strains in our collection. We used 375 additional S. suis genomes downloaded from the NCBI’s Sequence Read Archive (SRA) to validate the pipeline. Validation with SRA WGS data resulted in a 92 % match. Included pipeline subroutines permitted us to assess strain virulence marker content and obtain multilocus sequence typing directly from WGS data.ConclusionsOur pipeline permits rapid and accurate determination of S. suis serotype, and other lineage information, directly from WGS data. By discriminating between serotypes 1 and 14, and between serotypes 2 and 1/2, our approach solves a three-decade longstanding S. suis typing issue.

【 授权许可】

CC BY   
© The Author(s). 2016