323. Comparing Two Whole Genome Sequencing Bioinformatic Software for Identifying Enterococcal Antibiotic-Resistant Genes

Abstract Background Enterococci are leading causes of infections, and many strains are multidrug-resistant. Identification of antibiotic resistance can be achieved either phenotypically using Clinical & Laboratory Standards Institute (CLSI) minimum inhibitory concentration (MIC) cut-offs, or genotypically using whole genome sequencing (WGS). Here we compare the output from two software pipelines: EPISEQ CSTM (BIOMÉRIEUX, Marcy l ‘Etoile, France) and ResFinder to understand the similarities and differences in their ability to identify antibiotic resistance genes. Methods We performed WGS on 89 clinical isolates of Enterococci (both E. faecalis and faecium) from two distinct tertiary care Detroit hospitals admitted to 16 intensive care units (ICU) and non-ICU wards between 2017-2019. WGS was performed using the NextSeq (Illumina Inc., CA) and analyzed using two bioinformatics pipelines mentioned above. The antibiotic resistance outputs were compared to identify both similarities and differences. Results There was a significant difference in the genetic mutation output from both software. EPISEQ was able to identify 13 & 15 different genetic mutations for Enterococcus faecalis and faecium respectively which were not identified by ResFinder (Table 1). There were 9 & 12 common genetic mutations for Enterococcus faecalis and faecium respectively which were identified by both software. ResFinder was able to identify only two different gene mutations which were not identified by EPISEQ. Table 1:Genetic mutations identified by EPISEQ and ResFinder arranged by resistance to its respective drug class. Conclusion The ability to identify different genetic mutations by both software products likely depends on databases used to determine the antibiotic resistant genes. EPISEQ CS uses 4 different databases (including ResFinder) as well as other proprietary databases making it more sensitive, thereby able to identify more genetic mutations compared to ResFinder alone. The costs for analyzing the WGS data through proprietary software is steep compared to open-source software. Standards metrics and methods for interpretations are needed before these methods can be adapted for clinical applications. Further studies are needed to understand the value of identifying these additional genes by comparing it to phenotypic susceptibilities. Disclosures Piyali Chatterjee, PhD, AHRQ Grant # 1R03HS027667-01: Grant/Research Support|AHRQ Grant # 1R03HS027667-01: Central Texas Veterans Health Care System Keith S. Kaye, MD, MPH, Allecra: Advisor/Consultant|GlaxoSmithKline plc.: Receiving symposia honoraria|GlaxoSmithKline plc.: GlaxoSmithKline plc.-sponsored study 212502|Merck: Advisor/Consultant|qpex: Advisor/Consultant|Shionogi: Grant/Research Support|Spero: Advisor/Consultant Chetan Jinadatha, MD, MPH, AHRQ R01 Grant-5R01HS025598: Grant/Research Support|EOS Surfaces: Copper Coupons and materials for testing.