IdentifyingStreptococcus pneumoniaegenes associated with invasive disease using pangenome-based whole genome sequence typing

Abstract

AbstractStreptococcus pneumoniaeis a normal commensal of the upper respiratory tract but can also invade the bloodstream or CSF (cerebrospinal fluid), causing invasive pneumococcal disease (IPD). In this study, we attempt to identify genes associated with IPD by applying a random forest machine-learning algorithm to whole genome sequence (WGS) data. We find 43 genes consistently associated with IPD across three geographically distinct WGS data sets of pneumococcal carriage isolates. Of these genes, 23 genes have previously shown to be directly relevant to IPD, while the other 18 are uncharacterized.