Pseudomonas aeruginosa is a ubiquitous microorganism and an important opportunistic
pathogen responsible for a broad spectrum of infections mainly in immunosuppressed
and critically ill patients. Molecular investigations traditionally rely on pulsed
field gel electrophoresis (PFGE) and multilocus sequence typing (MLST). In this work
we propose a core genome multilocus sequence typing (cgMLST) scheme for P. aeruginosa,
a methodology that combines traditional MLST principles with whole genome sequencing
data. All publicly available complete P. aeruginosa genomes, representing the diversity
of this species, were used to establish a cgMLST scheme targeting 2,653 genes. The
scheme was then tested using genomes available at contig, chromosome and scaffold
levels. The proposed cgMLST scheme for P. aeruginosa typed over 99% (2,314/2,325)
of the genomes available for this study considering at least 95% of the cgMLST target
genes present. The absence of a certain number gene targets at the threshold considered
for both the creation and validation steps due to low genome sequence quality is possibly
the main reason for this result. The cgMLST scheme was compared with previously published
whole genome single nucleotide polymorphism analysis for the characterization of the
population structure of the epidemic clone ST235 and results were highly similar.
In order to evaluate the typing resolution of the proposed scheme, collections of
isolates belonging to two important STs associated with cystic fibrosis, ST146 and
ST274, were typed using this scheme, and ST235 isolates associated with an outbreak
were evaluated. Besides confirming the relatedness of all the isolates, earlier determined
by MLST, the higher resolution of cgMLST denotes that it may be suitable for surveillance
programs, overcoming possible shortcomings of classical MLST. The proposed scheme
is publicly available at: https://github.com/BioinformaticsHIAEMolecularMicrobiology/cgMLST-Pseudomonas-aeruginosa.
