826. Comparative Epidemiological Analysis of Serratia marcescens using PFGE and Whole Genome Sequence Methods

Abstract

BackgroundEpidemiological (EPI) analyses of bacterial pathogens play an important role in infection control practices during suspected outbreaks. Pulsed field gel electrophoresis (PFGE) is the gold standard for EPI typing of most bacterial organisms, but this method has been slowly replaced by sequencing-based methods, such as multilocus sequence typing (MLST) and core genome (cg)MLST that uses whole genome sequencing (WGS) data. We evaluated the utility of WGS ad hoc schemas to predict relatedness among Serratia marcescens (SM) clinical strains.MethodsA total of 19 SM clinical isolates collected as part of the SENTRY Program and JMI worldwide collections were selected. Isolates were typed by PFGE and analyzed using GelCompar II (100% similarity scored as identical, > 85% and < 100% as genetically related and < 85% as unrelated). WGS was performed using MiSeq (Illumina) and contigs generated using SPAdes. Raw reads and assembled contigs were used to generate phylogenetic trees using kWIP and progressiveMauve (pMauve), respectively. Similarity matrices and dendrograms created by the three protocols were compared.ResultsBased on PFGE analysis, 19 isolates were classified into 10 pulsotypes, A through J, and 1 subtype, A1 (Fig.1A). Among dendrograms generated based on WGS, analysis of short k-mers (Fig.1B) inaccurately showed phylogenetic separation among isolates in A/A1, C and G types, while analysis of much longer contigs (Fig.1C) accurately clustered isolates according to groups defined using PFGE. One isolate in the A/A1 group (517323) was separated from the group using kWIP; however, using the more sophisticated pMauve, this isolate was accurately clustered within the A/A1 isolates.ConclusionConcordance was observed between PFGE- and WGS-based phylogenetic analysis of SM for genetic relatedness. Based on this analysis, WGS data can be used to predict EPI of this bacterial species on an ad hoc basis. This methodology can be expanded to other species.Figure 1 DisclosuresRodrigo E. Mendes, PhD, A. Menarini Industrie Farmaceutiche Riunite S.R.L. (Research Grant or Support)Allergan (Research Grant or Support)Allergan (Research Grant or Support)Basilea Pharmaceutica International, Ltd (Research Grant or Support)Cipla Ltd. (Research Grant or Support)Department of Health and Human Services (Research Grant or Support)GlaxoSmithKline (Research Grant or Support)Melinta Therapeutics, Inc. (Research Grant or Support)Merck (Research Grant or Support)Merck (Research Grant or Support)Pfizer (Research Grant or Support) Mariana Castanheira, PhD, 1928 Diagnostics (Research Grant or Support)A. Menarini Industrie Farmaceutiche Riunite S.R.L. (Research Grant or Support)Allergan (Research Grant or Support)Allergan (Research Grant or Support)Amplyx Pharmaceuticals (Research Grant or Support)Cidara Therapeutics (Research Grant or Support)Cidara Therapeutics (Research Grant or Support)Cipla Ltd. (Research Grant or Support)Cipla Ltd. (Research Grant or Support)Fox Chase Chemical Diversity Center (Research Grant or Support)GlaxoSmithKline (Research Grant or Support)Melinta Therapeutics, Inc. (Research Grant or Support)Melinta Therapeutics, Inc. (Research Grant or Support)Melinta Therapeutics, Inc. (Research Grant or Support)Merck (Research Grant or Support)Merck (Research Grant or Support)Merck & Co, Inc. (Research Grant or Support)Merck & Co, Inc. (Research Grant or Support)Paratek Pharma, LLC (Research Grant or Support)Pfizer (Research Grant or Support)Qpex Biopharma (Research Grant or Support)