学术文献库

Objectives: To enhance monitoring of high-burden foodborne pathogens, there is opportunity to combine pangenome data with network analysis. Methods: Salmonella enterica subspecies Enterica serovar Enteritidis isolates were referred to the New South Wales (NSW) Enteric Reference Laboratory between August 2015 and December 2019 (1033 isolates in total), inclusive of a confirmed outbreak. All isolates underwent whole genome sequencing. Distances between genomes were quantified by in silico MLVA as well as core SNPs, which informed construction of undirected networks. Prevalence-centrality spaces were generated from the undirected networks. Components on the undirected SNP network were considered alongside a phylogenetic tree representation. Results: Outbreak isolates were identifiable as distinct components on the MLVA and SNP networks. The MLVA network based centrality/prevalence space did not delineate the outbreak, whereas the outbreak was clearly delineated in the SNP network based centrality/prevalence space. Components on the undirected SNP network showed a high concordance to the SNP clusters based on phylogenetic analysis. Conclusions: Bacterial whole genome data in network based analysis can improve the resolution of population analysis. High concordance of network components and SNP clusters is promising for rapid population analyses of foodborne Salmonella spp. due to the low overhead of network analysis.