Lab Matters Fall 2024 | Page 94

APHL 2024 POSTER ABSTRACTS

by screening all GenBank B . multivorans assemblies for the ST659 allelic profile . All data were assembled and analyzed at CDPHE . We used WGS-based multi-locus sequence typing ( MLST ) to determine sequence type as a method to quickly assess whether new potential cases could be ruled out as a part of the cluster . We performed further genomic analysis on the ST659 isolates to assess the extent of relatedness . Specifically , we performed a reference-free SNP analysis using kSNP3 and reference-based SNP analysis using Snippy , with predicted regions of genomic recombination masked using Gubbins . SNP alignments from kSNP3 and Snippy were used to generate phylogenetic trees . We used Microreact to overlay epidemiological data and sample metadata on the phylogenetic tree visualization .

Results : The ST659 isolates from the two states were genomically highly similar to each other , with between 0-17 SNP differences for kSNP3 ( mean 3.3 ) and between 0-14 for Snippy ( mean 2.4 ), when aligned to a ST659 reference genome . Inspection of phylogenetic tree structure did not reveal any indication of distinct clusters of isolates based on geographic location . This demonstrates that isolates from different states were approximately equally related to isolates from the same state . The epidemiological investigation revealed that the same brand of ice-water machine , cleaning products and water filters were used in the two states .

Conclusions : Epidemiological and genomic data suggests a possible common source of the outbreaks involving the ice machines . This investigation demonstrates the importance of coordinated analysis of genomic and epidemiological data between states to identify and resolve multi-state outbreaks .

Presenter : Samuel Baird , sam . baird @ state . co . us

Bio-Khoj : Coordinated Sentinel Surveillance and Discovery for Emerging Human Pathogens

S . Courtney 1 , J . Lee 1 , P . Rao 1 , A . Gundlapalli 1 , M . Mauldin 1 , J . Andreadis 1 , J . Schiffer 1 , C . Quiner 2 , K . Stolka 2 , C . Ferdon 1 , Centers for Disease Control and Prevention 1 , RTI 2

Ongoing detection and surveillance ( or khoj , meaning “ search ” in Hindi ) of emerging and reemerging human pathogens is central to public health outbreak preparedness . However , current diagnostic methods can be inadequate and the causative agent of many cases of suspected infectious diseases is often never identified . Efforts to use advanced molecular methods ( e . g ., “ shotgun metagenomics ”) to analyze routinely collected clinical specimens from people with undiagnosed severe illness are currently underway by CDC , public health and commercial and clinical laboratories , but are not sufficiently harmonized . Developing a coordinated , nationwide , threat agnostic surveillance program would provide a unique opportunity to understand the US landscape , lay the foundation for systematic emerging pathogen discovery at an unprecedented scale and significantly strengthen proactive preparedness and response capabilities .

Presenter : Sean Courtney , xcz1 @ cdc . gov

Carbapenemase-producing Organism Whole Genome Sequencing in the Antimicrobial Resistance Laboratory Network Northeast Region : Laboratory Information Management System Integration and Data Visualization

C . Prussing 1 , K . Cummings 1 , S . Morris 1 , D . Akerley 1 , Q . Brady 1 , C . MacGowan 1 , J . Levin 1 , A . Mehta 2 , K . Allen 3 , K . Musser 1 , E . Nazarian 1 , Wadsworth Center , New York State Department of Health 1 , NYS Office of Information Technology Services 2 , Center for Genomic Pathogen Surveillance / Digital Epidemiology Services 3

Managing bioinformatic pipeline data within a laboratory information management system ( LIMS ) is often a challenge , as these systems are not typically built for genomic data . Another key challenge is integrating bioinformatic pipeline data with demographic data and conventional laboratory results in ways that allow for detailed analysis and visualization , so trends can be tracked and viewed in real time . As the Northeast Regional laboratory for the Antimicrobial Resistance ( AR ) Laboratory Network ( AR Lab Network ), the Bacterial Diseases Laboratory at the Wadsworth Center ( WC ) has been conducting whole genome sequencing ( WGS ) on carbapenemaseproducing organisms ( CPOs ) since 2017 .

WGS data are analyzed using an in-house developed bioinformatics pipeline , the AR pipeline , which assigns multi-locus sequence type ( MLST ) and identifies AR genes present in the bacterial genome to the gene variant level . The pipeline is run on all AR isolates that undergo WGS , over 3,000 isolates to date and was approved in 2022 by NY state ’ s Clinical Laboratory Evaluation Program to be used for clinical reporting . Sequencing read files are uploaded to a Google Cloud Platform ( GCP ) storage bucket by laboratory staff , which triggers automated running of the bioinformatics pipeline . Detailed output , including logs and genome assemblies , are archived on GCP . Essential output fields including pipeline version , sequencing read file names , quality control QC metrics , MLST and AR genes identified are transferred to local servers and uploaded to WC ’ s clinical laboratory information management system ( CLIMS ).

Collection of the AR pipeline output data in CLIMS allows for electronic reporting of WGS results to both submitters and epidemiologists . Furthermore , AR pipeline data can now be pulled for multiple isolates in aggregate for reporting through a newly developed investigation module built in the CLIMS environment , used to report results from dozens of investigations to date . As pipeline output data are stored in CLIMS along with demographic data and other laboratory testing data for each isolate , it is possible to efficiently query and summarize these data for all isolates over time to answer epidemiologic and surveillance questions . We are now in the process of working with Digital Epidemiology Services and the Center for Genomic Pathogen Surveillance to develop a dashboard using Data-flo and Microreact software , to enable visualization of real-time trends in MLST types , AR genes and other key pieces of information extracted from CLIMS . The software tools transform , summarize and visualize these data to enhance data interaction and exploration and to streamline interpretation .

Integrating outputs of bioinformatics pipelines into existing data systems and designing efficient ways to summarize and view that data in real-time is a common challenge across laboratories . Through the integration of the AR pipeline output into CLIMS and