food safety and inspection service - aphl€¦ · office of public health science. food safety...

Food Safety and Inspection Service:Food Safety and Inspection Service:

1

Food Safety and Inspection Service:

Managing the Transition to WGS and Maintaining Multiple Workflows in

USDA FSIS

Glenn E. Tillman, PhD

Chief, Microbiology Characterization BranchEastern Laboratory (Athens, GA)Office of Public Health Science

Food Safety Inspection Service, USDASeptember 26-28, 2018

2


3

Why WGS?

• Improved resolution for foodborne illness investigations

• Supports FSIS mission goals– Understand foodborne illness and emerging microbiological trends– Recurrences of pathogens in FSIS-regulated establishments/products

to further support the inspection and verification process

• Alignment of pathogen surveillance with our domestic public health and regulatory partners

PresenterPresentation NotesWhy is FSIS implementing WGS analyses?

Improved discrimination – outbreaks, defining cases

Mission goals – incorporating new technologies to help us understand relationship between food products and illness and to look at other emerging microbiological trends

Aligning with our public health and regulatory partners in the US and abroad


• Performs WGS on all isolates from FSIS sampling programs– 12 sequencers in FSIS Field Service Laboratories– In FY18, FSIS has sequenced ~11,000 isolates

• Use WGS analyses in addition to epidemiological and traceback information to further understand the relationship between isolates

• Work with National Antimicrobial Resistance Monitoring System (NARMS) partners (FDA, CDC) to understand the occurrence or introduction of antimicrobial resistance genes in pathogens of interest

4

WGS at FSIS: Current Status

PresenterPresentation NotesFSIS usually has roughly 8000-10000 isolates per year from our various testing programs – the intent is to sequence roughly half of those isolates in a real-time manner.

We are currently using WGS data as part of our outbreak investigations and other harborage related analyses for Lm, particularly in DJE.


5

1- FSIS Eastern Lab began WGS on adulterant isolates2- Uploaded first Lm sequences to NCBI

FY2014

1- Visits to NCBI, FDA and CDC2- Procurement of two MiSeq’s3- First Sequences produced for outbreak investigations

FSIS shared Lmwith FDAfor WGS


FY2015

FSIS begins WGS on selected cecalSalmonella and Campylobacter

FY2016

1- WGS on all Salmonella and Campylobacter2- FSIS Midwestern Lab begins WGS

FY2017

1- FSIS Western Lab begins WGS2- PFGE no longer performed on Lm3- WGS on NARMS cecal E. coli and Enterococcus4- Use WGS for reporting in Lm

Food Safety and Inspection ServiceWGS at FSIS: Capability and Capacity Building by Year

FY2013 FY2018

FY2019

1- Stop PFGE for Campylobacter2- Stop PFGE for Salmonella & STEC


6

WGS Data Sharing – Metadata and Sequence Data

• Description about an isolate that has an experiment assigned • Metadata, such as source type, organism, serotype etc.• BioSample number is specific to a bacterial isolate for FSIS

FSIS Submissions to NCBI Bioprojects• PRJNA242847

– GenomeTrakr Project: USDA-FSIS (Salmonella)

• PRJNA215355– GenomeTrakr Project: FDA (Listeria

monocytogenes)

• PRJNA287430– USDA-FSIS: Campylobacter

• PRJNA268206– GenomeTrakr Project: USDA-FSIS (STEC)

• PRJNA292666– FSIS NARMS Salmonella

• PRJNA292667 – FSIS NARMS E. coli

• PRJNA292668– FSIS NARMS Campylobacter

• PRJNA292669– FSIS NARMS Enterococcus

PresenterPresentation NotesImportant points: we release minimal metadata, including year of isolation and state where the sample was collected in “real-time” – which means that when the sequence is available for upload, so is the metadata. We upload sequences whenever they are run – there is no delay (including during investigations).


FSIS: WGS Data Analyses Work Flow Overview

7

Output: FASTA

Input: FASTQ

• MLST Sequence Type• Antibiotic Resistance genes• Virulence Profile• Salmonella and STEC serotype• MASH Tree comparison

@M02848:54:000000000-AJ347:1:1101:14587:1926 1:N:0:4 TCCGTGCTCAGTTACACGGACAAAATACCGGCGAAAAACCTTGGGCCTTCCCTGGCGACATGGGATT + AA>1A1AAD31DF3F3B111A0000B1BB00A/////0AAGG11/BFFGFFFG1/>//E/?10E0FGDGFG1

• wgMLST BioNumerics 7.6• Lyve-SET, SNP Pipeline• NCBI Pathogen Isolate Browser

De novo Assembly

Input: FASTA

QC Pipeline• Coverage• Average Quality• Nucleotide balance

QC Pipeline• File Size• N50 & No.

contigs• Correct organism


8

Maintaining Parallel Workflows

• Four main characterization tests currently performed in parallel

0

2000

4000

6000

8000

10000

12000

14000

16000

PFGE AST WGS Serotyping

Bacterial Isolates Characterized by Test

2014 2015 2016 2017


Further Characterization of Isolates Using WGS

• Campylobacter speciation • Serotyping/serogrouping

– Salmonella– Adulterant STEC

• Antimicrobial Resistance (Phenotype prediction)

– Salmonella– Campylobacter– E. coli– Enterococcus

• Identify characterized genes of interest– Resistance to environmental factors (heat, acid,

sanitizers, etc)– Virulence factors

• stx/eae sub-types (STEC)• Alternative to PFGE for comparison of

genotypes– wgMLST analyses– SNP analyses

A single workflow for many characterization approaches viainformatics

9

PresenterPresentation NotesIn addition to surveillance and regulatory purposes, WGS has the potential to replace a number of subtyping analyses, leading to a more efficient workflow for characterization of isolates.

In most cases, WGS analyses will take more time than traditional methods performed individually, but all of the information can be obtained from a single characterization technique (WGS with informatics).


10

WGS Analyses for Phylogenetic Context: wgMLST and hqSNP Analyses

FSIS uses pipelines developed by public health partnersLyve-SEThttps://github.com/lskatz/lyve-SET/blob/master/

NCBI Pathogen Detection Isolates Browswerhttp://www.ncbi.nlm.nih.gov/pathogens

FDA SNP Pipelinehttps://github.com/CFSAN-Biostatistics/snp-pipeline

wgMLSTBioNumerics 7.6 CDC-PulseNet

https://github.com/lskatz/lyve-SET/blob/master/http://www.ncbi.nlm.nih.gov/pathogens


11

Applications of WGS at FSIS: Lm HarborageFood Safety and Inspection ServiceWGS usage: hqSNP analyses and wgMLST with Listeria monocytogenes

Lyve-SETSnp Pipeline

wgMLST

0-13 SNPs

NCBI Pathogen Browser hqSNP

PresenterPresentation NotesWGS and PFGE results are generally in agreementSame PFGE pattern: WGS agreement0-2 SNPs0-7 allele differences

WGS can sometimes exclude isolates from a group within the same PFGE pattern Same PFGE pattern: WGS exclusion 34-45 SNPs 29-33 allele differencesor include isolates in a group with different PFGE patterns (wgs has finer reading capability)Different PFGE: WGS inclusion 0-3 SNPs 0-5 allele differences

In the examples below inclusion or exclusion considered all information including sample metadata (establishment, isolation date, etc.) CDC wgMSLTFDA uses high quality SNPs


• High-throughput capacity for WGS – Sequenced ~11,000 isolates in FY18 to date

• FSIS has performed parallel workflows for several years

• FSIS labs will utilize WGS data in outbreak investigations, and as a stream-lined analytical procedure

12

Concluding Remarks


Acknowledgements

13

• USDA FSIS Offices• USDA ARS• CDC PulseNet and NARMS• FDA CFSAN• FDA CVM• NCBI• State Laboratories

Slide Number 1Managing the Transition to WGS and Maintaining Multiple Workflows in USDA FSISWhy WGS?WGS at FSIS: Current StatusSlide Number 5WGS Data Sharing – Metadata and Sequence DataFSIS: WGS Data Analyses Work Flow OverviewMaintaining Parallel WorkflowsFurther Characterization of Isolates Using WGSWGS Analyses for Phylogenetic Context: wgMLST and hqSNP AnalysesApplications of WGS at FSIS: Lm HarborageConcluding RemarksAcknowledgements

food safety and inspection service - aphl€¦ · office of public health science. food safety...

Documents