SeroBA: rapid high-throughput serotyping of Streptococcus pneumoniae from whole genome sequence data

Area of Science:

• Microbiology
• Genomics
• Bioinformatics

Background:

• Streptococcus pneumoniae causes significant child mortality globally.
• Accurate serotyping is crucial for monitoring vaccine impact and pathogen evolution.
• Existing genomic serotyping tools have limitations in scalability and performance.

Purpose of the Study:

• To introduce SeroBA, a novel k-mer based method for rapid and accurate pneumococcal serotype identification.
• To evaluate SeroBA's performance, scalability, and robustness using real and simulated genomic data.
• To enable efficient serotype prediction directly from raw whole genome sequencing reads.

Main Methods:

• Development of SeroBA, a k-mer based bioinformatics tool.
• Comparative analysis of SeroBA against existing methods using validation and large datasets.
• Assessment of SeroBA's performance across varying whole genome sequencing coverage depths.

Main Results:

• SeroBA achieves 98% concordance in serotype prediction by identifying the cps locus.
• The method demonstrates high scalability, processing 10,000 samples in approximately one day.
• Accurate serotyping is achievable with low sequence coverage (15-21×).

Conclusions:

• SeroBA offers a robust, scalable, and accurate solution for pneumococcal serotype identification from WGS data.
• The tool facilitates efficient genomic epidemiology and surveillance of Streptococcus pneumoniae.
• SeroBA is freely available as an open-source Python3 package.