Lab Matters Fall 2024 | Page 98

APHL 2024 POSTER ABSTRACTS

Results : 15 of the 16 SNP calls and all three allele calls showed 100 % accuracy , specificity and sensitivity for both versions of the pipeline . The two sequence typing methods also showed 100 % accuracy , specificity and sensitivity , once we updated and aligned the databases for both versions of the pipeline . One SNP call showed 98 % accuracy , 97 % sensitivity , but 100 % specificity .

Conclusions : Our preliminary findings illustrate that this protocol and dataset are suited for assessing this custom Ng AMR analysis pipeline . Further work will increase the reference and test data set to include relevant sequences for validation of additional variants of interest . Validation of this pipeline provides a reliable , standardized approach for analyzing genomic data from Ng isolates and provides a framework for development of similar validation sets for a range of pathogens using the NGS QI protocols .

Presenter : Katherine Hebrank , uce1 @ cdc . gov

Enhancing Whole Genome Sequencing Accuracy and Consistency through External Quality Assessment in the Surveillance of Neisseria gonorrhoeae

J . Reimche , A . Smith , C . Pham , M . Schmerer , J . Cartee , C . Bolden , E . Kersh , K . Gernert , Centers for Disease Control and Prevention

Introduction : Drug-resistant Neisseria gonorrhoeae ( Ng ) poses an urgent threat , as classified by the CDC . Preserving and optimizing existing antibiotics requires a robust Ng surveillance network to monitor antimicrobial resistance ( AR ) and strain types through whole genome sequencing ( WGS ). With the increasing accessibility of WGS , the need for quality control metrics to ensure consistent , highquality data across laboratories remains crucial . In response , we implemented an external quality assessment ( EQA ) program in the Fall of 2020 within the Antimicrobial Resistance Laboratory Network ( AR Lab Network ) to biannually monitor WGS .

Methods : A panel of five blinded reference Ng isolates was distributed in triplicate to four regional labs for antimicrobial susceptibility testing ( AST ) and Illumina MiSeq WGS following standard protocols . The generated reads and run data were sent to the CDC STD Laboratory Reference and Research Branch ( SLRRB ) for taxonomic classification , assembly , phylogenetics and analysis . Quality evaluation included % Q30 , expected coverage , total aligned length , % Ng reads out of all Neisseria , read depth of coverage , number of contigs , N50 , Multilocus Sequence Type ( MLST ) and alignment SNP distance .

Results : Five laboratories participated in a pilot and five EQA cycles from Spring 2021 to Spring 2023 . In the pilot cycle , all labs achieved 100 % across all quality metrics . However , in Spring 2021 , one lab scored 66.7 %, revealing the importance of ongoing assessment . Failed specimens showed low expected coverage and low-quality assemblies . MLST and SNP distances proved valuable in identifying specimen swapping errors . In Spring 2022 , two labs failed with < 75 % pass rates , uncovering concerns in specimen handling in one lab and low coverage and low % Ng reads in another . Following root cause analysis and corrective actions , all labs achieved 100 % pass rates for Fall 2022 and Spring 2023 .

Conclusions : The Neisseria gonorrhoeae WGS EQA program identified weaknesses in laboratory WGS workflows and revealed sample management errors unlikely to be discovered during routine sample processing . A quality management tool which aligns phenotypic ( AST ) and genotypic data confirms accurate specimen / sequence alignment for regular monitoring . Addressing these errors enhances data quality for routine sequencing in participating labs and ensures WGS consistency across the participating Ng AR Labs . This comprehensive approach contributes to the effective surveillance of drug-resistant Neisseria gonorrhoeae , identification of genomic variants of AMR and supports the sustainable use of antibiotics .

Presenter : Amanda Smith , rqq8 @ cdc . gov

Establishment of Prospective Genomic Surveillance and Characterization of Respiratory Syncytial Virus in Minnesota

D . Evans 1 , H . Kunerth 1 , S . Namugenyi 1 , M . Plumb 1 , S . Bistodeau 1 , E . Mumm 1 , B . Schmitt 2 , S . Cunningham 1 , K . Como-Sabetti 1 , R . Lynfield 1 , S . Wang 1 , Minnesota Department of Health 1 , Children ’ s Minnesota 2

Background : Respiratory syncytial virus ( RSV ) is a major respiratory pathogen . RSV surveillance using whole-genome sequencing ( WGS ) technology may strengthen our understanding of transmission dynamics and virulence . Our objective was to initiate genomic surveillance of RSV infections in Minnesota to track circulating viral subtypes and mutations , as well as to identify outbreaks from genetic clusters .

Methods : Positive RSV respiratory specimens from outpatient and inpatient cases were sent from six healthcare facilities to the Minnesota Public Health Laboratory for WGS . All genomes from viral specimens were amplified using tiled amplicon-based primers and sequenced using the GridION platform by Oxford Nanopore Technologies . Genome assembly , quality control and viral subtyping were performed using the nf-core Viralrecon pipeline using custom primer schemes and Nextclade lineage designations . Annotated phylogenetic trees were constructed for RSV subtypes A and B using a modified version of the publicly available Nextstrain build for RSV , based on the Augur bioinformatics software . Single nucleotide polymorphisms ( SNPs ) were called from MAFFT-based Augur alignments using snp-dists v0.8.2 . Phylogenetics results and metadata were regularly shared with the Minnesota RSV-NET program to link pathogen genomic data to epidemiologic data from hospitalized patients .

Results : From August to December 2023 , 275 RSV genomes were successfully sequenced , assembled , analyzed and cross-referenced against epidemiologic data . Of these genomes , 147 ( 53.5 %) were classified as RSV-A and 128 ( 46.5 %) were classified as RSV-B . Using our subtype-agnostic sequencing and bioinformatics protocol , we identified one coinfection of both subtypes in the same case . RSV-A genomes were distributed among four phylogenetic clades , with the majority in two clades of the A . D . 5 lineage ( n = 109 , 74.1 %). By contrast , nearly all RSV-B genomes belonged to a single clade and lineage , B . D . E . 1 ( n = 122 , 95.3 %). 61 ( 41.5 %) RSV-A genomes and 39 ( 30.5 %) RSV-B genomes were indistinguishable from at least one other sequence from another case . RSV-B genomes taxonomically classified within the B . D . E . 1 lineage exhibited greater genetic diversity than that observed among RSV-A genomes of similarly specific lineage classifications . Additionally , RSV-A genomes exhibited trends of closer overall genetic relatedness within the subtype than was observed among RSV-B genomes . Genomic