11/28/2017 Whole Genome Sequencing for Cluster Detection Minnesota, - - PDF document

11/28/2017 1

Whole Genome Sequencing for Cluster Detection Minnesota, 2017

Carlota Medus, PhD, MPH Epidemiologist Supervisor Sr. Foodborne Diseases Unit

WGS for Cluster Detection in Minnesota

• Whole genome sequencing (WGS) for S. Enteritidis (SE) cluster

detection since 2014

• WGS for S. Typhimurium (STm) cluster detection since 2017

−SE and STm isolates sequenced at MDH −hqSNP analysis at Wadsworth Center, New York State Department of Health

• All other Salmonella and STEC are sequenced, but only analyzed
• n request. All Listeria isolates are also sequenced at MDH and

analyzed at CDC

• Subtyping using pulsed‐field gel electrophoresis (PFGE) continues

SE and STm WGS Data Flow

Sequenced at MDH Illumina Basespace Wadsworth Center, NYDOH

hqSNP analysis

(trees, heat maps, raw data files)

MDH Lab

evaluates the data, identifies clusters

Notification to epi

11/28/2017 2

• Includes closely

related (within 10 SNPs) 2014‐ 2017 isolates

0.02

SE hqSNP Tree, April 2017

11/28/2017 3

10 9 8 7 6 5 4 3 2 1 1 2 3 4 5 6 7 8 9 10

WGS for Cluster Detection

• Define “cluster”

−Specificity of case definition is important for finding the epi link −Flexibility to later expand the case definition if warranted by the epi data

MDH00219 MDH00225‐In‐vivo, same as E2001001070 MDH00215 ‐Sporadic 4/19/01 MDH00247 ‐‐Sporadic 8/6/12 MDH00204 ‐ Sporadic 5/14/01 MDH00221‐ Sporadic 5/14/01 MDH00203 ‐ Sporadic 7/11/00 MDH00214 ‐ Sporadic 3/12/01 MDH00206 ‐ Sporadic 8/23/00 MDH00217 ‐ Sporadic 6/10/13 MDH00237 Sporadic 6/22/11 MDH00236 ‐ Sporadic 5/7/11 MDH00207 ‐ Sporadic 8/31/2000 MDH00233 ‐ Sporadic 12/7/2001 MDH00248 ‐ Sporadic 6/10/13 MDH00205 ‐ Sporadic 8/22/2000 MDH00216 ‐ Sporadic 4/30/2001 MDH00224 ‐Sporadic 6/11/2001 MDH00254 MDH00252 MDH00253 MDH00234 MDH00226 ‐ Sporadic 6/21/2001 MDH00231 ‐ Sporadic 7/16/2001 MDH00202 ‐ Sporadic 7/7/2000 MDH00208 ‐ Sporadic‐ Same time, PFGE, and MLVA as Outbreak 1 MDH00209 MDH00210 MDH00211 MDH00222‐In‐vivo, same as E2001001070 MDH00228‐In‐vivo, same as E2001001070 MDH00223 MDH00220 MDH00218 MDH00213‐Sporadic‐ Same PFGE and time as Outbreak 1 MDH00232‐Sporadic 10/17/01 MDH00227 MDH00230 MDH00251 MDH00229 MDH00235‐Sporadic 10/3/05 MDH00243‐Sporadic, same PFGE and time as Outbreak 5 MDH00245‐Sporadic 6/26/12 MDH00249 MDH00250 MDH00246‐Sporadic 7/30/12 MDH00255‐OH Sample 1 MDH00256‐OH Sample 2 MDH00241‐Sporadic, same PFGE and time as Outbreak 5 MDH00239 MDH00242 MDH00244‐Environmental sample from Outbreak 5 MDH00238 MDH00240

Defined Outbreak Samples Outbreak 1‐ Sept 2000 Outbreak 2‐ May 2001 Outbreak 3‐ Aug 2001 Outbreak 4‐ Nov 2003 Outbreak 5‐ Aug 2008 Outbreak 6‐ Spring 2014 Outbreak 7‐ Spring 2014 0-2 SNPs 0-2 SNPs 0-2 SNPs 1 SNP 0 SNPs 0-1 SNPs 0-3 SNPs

Taylor et al. J Clin Micro Oct 2015

Salmonella Enteritidis Retrospective Study, Minnesota, 2000‐2014

11/28/2017 4

Lab ID PFGE pattern Number of SNPs Lab ID for closely related isolates Cluster ID

WGS Results Spreadsheet

Number of SNPs Lab ID for closely related isolates

WGS Results Spreadsheet

11/28/2017 5

Shared Network Folder Communication of Results, Summary

• Lab

−Interprets the analysis −Identifies clusters −Enters data in a spreadsheet −Sends email to epis

• Epi

−Rely on email to start cluster investigations −Have access to

• Spreadsheets
• hqSNP Trees
• Heat map (SNP matrix)

SE Cluster Comparison, January‐October, 2017

PFGE WGS Number of clusters 14 22 Median number of cases per cluster 3.5 3 Range of number of cases per cluster 2 ‐ 62 2 ‐ 13 Number of clusters with ≥10 cases 5 2 Number of clusters with pattern .0004 isolates 1 8 Number of clusters with pattern .0002 isolates 1 7

11/28/2017 6

Domestic WGS SE Clusters, January‐October, 2017

Cluster ID Corresponding PFGE Number of Cases Epidemiological Link 2 .0004 13 Kitfo from Ethiopian store subcluster; other 4 .0021 12 Restaurant 9 .0049 9 Likely lettuce (multi‐state) 16 .0004 8 Live poultry (multi‐state) 17 .0005 6 Live poultry (multi‐state) 7 .0004 4 Unsolved 19 .0004 & .0034 3 Live poultry (multi‐state) 21 .0004 3 Live poultry suspected 20 .0004 2 Live poultry suspected

Travel‐Associated WGS SE Clusters, January‐ October, 2017

Cluster ID Corresponding PFGE Number of Cases Epidemiological Link 1 .0002 7 Dominican Rep. 6 .0004 6 Mexico‐some same resort 10 .0002 4 Dominican Rep.‐some same resort 5 .0004 3 Jamaica‐same resort 3 .0008 2 Mexico‐same resort 11 .0116 & .1076 2 Dominican Rep. 12 .0002 2 Dominican Rep.‐same resort 13 .0054 2 Dominican Rep. 14 .0042 2 Mexico‐same resort 15 .0004 2 Mexico 18 .0002 2 Dominican Rep. No ID .0002 2 Dominican Rep. No ID .0002 2 Dominican Rep.

WGS for SE Cluster Detection

• Real‐life example of using WGS in surveillance

−January‐April, 2017

11/28/2017 7

0.02

• Isolates in a 90‐

day time period

• Includes closely

related (within 10 SNPs) 2014‐ 2017 isolates

2017 SE hqSNP Tree, April, 2017

Cluster 3 Cluster 6 Cluster 11 Cluster 13 Cluster 10 Cluster 12 Cluster 1

Travel to Dominican Republic Travel to Mexico

0.02 0.02

• Isolates in a 90‐

day time period

• Includes closely

related (within 10 SNPs) 2014‐ 2017 isolates

SE hqSNP Tree, April, 2017

11/28/2017 8

Cluster 9 (lettuce) Cluster 7 (unsolved) Cluster 5 Cluster 2 (kitfo) Cluster 4 (restaurant)

Travel to Jamaica PFGE pattern .0004

• No. of Cases

Week of Specimen Collection

Cases by week of specimen collection, PFGE alone (n=27)

Week of Specimen Collection

• No. of Cases

WGS Cluster cases by week of specimen collection

11/28/2017 9

Cluster 4

April, 2017 SE hqSNP 60‐day tree

• No. of Cases

Week of Specimen Collection

Cases by week of specimen collection, PFGE alone (n=17)

• No. of Cases

Week of Specimen Collection

Cases by week of specimen collection, PFGE alone (n=17)

Week of Specimen Collection

• No. of Cases

Cases by week of specimen collection by WGS

11/28/2017 10

Cluster 4

June 2017 SE hqSNP 60‐day tree

4‐6 SNP from the outbreak 13‐16 SNP from the outbreak hqSNP, June 2017 Cluster 4‐ Restaurant

• utbreak, 0‐2 SNPs

Part of the Outbreak or Not?

• Specimen collection 4/8/17, onset of illness 3/21
• Lives 35 miles from the outbreak restaurant

Outbreak isolates

11/28/2017 11

Part of the Outbreak or Not?

• Specimen collection 5/15 and 5/22
• (>2 months after last patron

specimen collection date)

• 167 miles from the outbreak

restaurant

• Both 17 years of age
• One not interviewed
• One did not eat at the restaurant

and… Outbreak isolates

11/28/2017 12

0.02

60‐day hqSNP Tree, September, 20XX

Hypothetical Example, Salmonella Baconitidis

Cluster 2 ‐ Vet Diagnostic Lab isolates; porcine Cluster 3 ‐ Restaurant outbreak; pork suspected Cluster 4 ‐ Wedding reception outbreak Cluster 5 ‐ Interviews pending Cluster 1 ‐ Outbreak associated with a pig roast at an event

0.02

Hypothetical Example, Salmonella Baconitidis

Previous year ‐ pork‐assoc. outbreak Cluster 2 ‐ Vet Diagnostic Lab isolates; porcine Cluster 3 ‐ Restaurant outbreak; pork suspected Cluster 4 ‐ Wedding reception outbreak Cluster 5 ‐ Interviews pending Cluster 1 ‐ Outbreak associated with a pig roast at an event

0.02

Hypothetical Example, Salmonella Baconitidis

Previous year ‐ pork‐assoc. outbreak

11/28/2017 13

Ongoing Issues

• Cluster approach generalizable to other pathogens, to other states

−Conceptually, yes:

• Specificity of the case definition; expand if necessary

−Some of the details, not:

• The precise number of SNPs will vary and the tree could look

different depending on how the analysis is done, the pipeline, the reference genome

• The number of SNPs we should consider closely related may

be different for different pathogens, and maybe even affected by the reservoir

Final Thoughts

• The power of WGS in epidemiology is readily

apparent −Better defines which cases are part of an outbreak

• Using WGS for cluster identification is not simple and
• ne‐directional

−i.e., “Related” sequences may or may not represent a common source outbreak; may differ by serotype −Epidemiology is extremely important in informing what we should consider a cluster

Final Thoughts (cont.)

• wgMLST

−New strategies for communication will have to be developed −Standardization of reports from lab to epi

11/28/2017 14

Final Thoughts (cont.)

• Epi‐lab collaboration and on‐going communication are

key

Acknowledgements

Minnesota Dept. of Health, Foodborne Diseases Unit Kirk Smith Dana Eikmeier Marijke Decuir Josh Rounds Amy Saupe Stephanie Meyer Team Diarrhea Federal Partners CDC FDA NCBI Minnesota Dept. of Health, Public Health Laboratory Angie Jones Taylor Dave Boxrud Xiong (Sean) Wang Megan Nichols Victoria Lappi Enterics lab, PFGE lab, & Sequencing Core New York Dept. of Health Wadsworth Center Bill Wolfgang Pascal Lapierre Kim Musser Samantha Wirth Bioinformatics Core

11/28/2017 15

Thank you

Carlota.Medus@state.mn.us

November 28, 2017

Communication Strategies for Laboratorians and Epidemiologists

• n NGS Data

Wadsworth Center, NYSDOH Experience

Kimberlee Musser, PhD Chief of Bacterial Diseases Wadsworth Center

How did we get here?

• Whole-genome sequencing ~ 5 years
• Numerous In-house Bioinformatics Pipelines

– Bacterial Foodborne Disease – Legionella – Clinical TB WGS

• Projects/Activities

– Wadsworth Center Public Health Genomics Initiative – FDA GenomeTrakr – CDC- PulseNet, AMD Legionella, Prevent Block – CDC/APHL pilot TB WGS – ELC- Lab Capacity and ARLN

11/28/2017 16

https://www.boardofinnovation.com

How do we communicate NGS/WGS results in an impactful way?

• Outbreaks/ Investigations
• Requests/ Applied Research
• Successes!

Can we try WGS? How many SNPs indicate relatedness? When will we have results? Have we seen this strain before?

Reporting to Epidemiology

• Modeling after PFGE reporting
• Discussion with bioinformaticians
• Trial and error evolving process
• Input
• What do we send? Comparison to whole

tree, specific time period, geographical, requested sample

11/28/2017 17

Reporting to Epidemiology

• Modeling after PFGE reporting
• Discussion with bioinformaticians
• Trial and error evolving process
• Input
• What do we send? Comparison to whole

tree, specific time period, geographical, requested sample

Text Trees Interpretations SNPs Heat maps Relatedness to… Recombination

Reporting to Epidemiology

• How do we send?
• Patient names, initials, Identifiers
• Type of analysis
• Who do we send to- Epi NYS, NYC, both,
• Env. Health, CDC

One size doesn’t fit all…

11/28/2017 18

Always consider Lab WGS Results in context of Epidemiology

https://www.coursera.org/learn/epidemiology

November 28, 2017 53

Foodborne Bacterial Disease

Communicating NGS for Foodborne Bacteria:

We communicate NGS results with a number of partners:

– CDC – Other state labs – FDA – NYS epidemiologists

• Unsolicited cluster reports
• Epi. solicits cluster analysis for specific samples

11/28/2017 19

Criteria for reporting WGS clusters

Organism Minimum # isolates in cluster SNP diversity in the cluster Sample collection date range for analysis Salmonella Enteritidis 3 0-5 60 days Salmonella Typhimurium 2 0-10 90 days STEC 2 0-20 365 days

• L. monocytogenes

2 0-30 alleles* No limit

• Routine testing
• SE most stringent criteria
• L. mono. least stringent criteria
• Other organism and serotype analysis performed by CDC
• Based on practical experience

BUT can and do change over time.

Daily communication

https://nysemail.sharepoint.com/sites/healthcch/EBCDC/Lists/WGS%20requests/AllItems.aspx

…a work in progress

Daily communication

https://nysemail.sharepoint.com/sites/healthcch/EBCDC/Lists/WGS%20requests/AllItems.aspx

…a work in progress

11/28/2017 20

• Criteria by organism for reporting WGS

clusters

• SharePoint (PFGE is SFT)
• NYS Epi

What, how, and who…

November 28, 2017 59

Mycobacterium tuberculosis

Bioinformatics

Kraken K-mer matching Detect spacers (allow 1 SNP) SNP calling with indels, major deletions SNP calling ignore indels Map to Reference Genome

1 2 3 4

MTBC member ID Spoligotyping Prediction of antibiotic resistance Build consensus for phylogeny

Report

TB Whole Genome Sequencing and Pipeline Overview

11/28/2017 21

• TB WGS phylogenetic tree is constructed every

week, or as requested by TB control (NYS or NYC)

• The tree currently takes ~5 hours to run as all

sequences in our database are assessed (1500)

• The time period for relevant strain matches can

span years (not weeks).

Phylogenetic analysis TB WGS tree

Euro-American (53%) Beijing (23%) CAS (6.3%) EAI (13.3%) Other species (4.4%)

Reporting Related TB strains

We report to TB Control on any isolates that have a low number of SNPs (<20 SNPs). Reporting includes:

• Spoligotype
• Isolate IDR numbers/ submitter number if applicable
• Brief demographics (age, gender, collection date, submitter,

county).

• SNPs, heat map, screenshot of tree
• Contaminating bacterial DNA?

11/28/2017 22

Example 1:

• M. bovis strains w/ SM resistance – S00172

IDR1500057177 47M unknown county, sputum collected 7/29/15 from Lab A IDR1600011196 33M Kings, sputum collected 2/4/16 from Lab B IDR1600055557 4M Kings, bone of the foot collected 11/9/16 from Lab B IDR1700025819 71M Kings, sputum collected 3/15/17 from Lab B 0‐5 SNPs difference – Patients may have been infected from an unpasteurized cheese product (Queso Fresco)

S00388 NYS County Hotel Outbreak IDR1400004233-01-02 37M County A sputum collected 2/6/14 IDR1400009529-01-01 37M County A sputum collected 2/6/14 IDR1600012456-01-02 59F County A sputum collected 3/15/16 IDR1600026772-01-01 32M County A sputum collected 5/21/16 IDR1600020177-01-01 58M County A sputum collected 4/11/16 IDR1600039993-01-01 28M County B sputum collected 8/11/16 0-1 SNP differences These strains are related to an ongoing outbreak that originated at County A convention in a hotel.

Example 2:

Employees Unclear/unknown link Index case

• Any match in entire tree that is <20

SNPs

• Email/ phone (Patient initials)
• NYS and/or NYC TB Control

What, how, and who…

11/28/2017 23

November 28, 2017 67

Legionella

Legionella Outbreak Communication

Lab HAI Epi BCDC Epi Environmental Health Legionella Case Reports

Interviews Medical record reviews Interaction with facility, physicians Community or HAI Request cultures, specimens Assessment of facility Interaction with facility Review of maintenance records Sampling recommendations Discussing testing volumes, types Preparing – ordering media, labelling Testing timelines Shipping, Testing data‐ WGS

📲📟📨 📲📟📨 📲📟📨

Regional Epi

WC received specimens almost every day for 3 weeks

Cooling Tower Potable Water NYC NYS Isolates Specimens

31 cases 44 cases 65 cases, 1 death 81 cases, 7 deaths 100 cases, 10 deaths 138 cases, 16 deaths

90% cases hospitalized

11/28/2017 24

Autopsy lung culture Cooling tower culture Real‐time PCR Cooling tower sampling Cooling tower Processing

• ST. BARNABAS NURSING HOME
• ST. BARNABAS HOSPITAL
• ST. BARNABAS HOSPITAL NURSING.
• ST. BARNABAS NURSING HOME
• ST. BARNABAS NURSING HOME

PFGE

<24 hours ~ 14 days 4‐5 days

Company Tower

Hotel H Hotel H Towers Hospital A Hospital L

Plaza

Company C

>13,500 SNPs

0‐3 SNPs

>170,000 SNPs

South Bronx clinical

80 samples (clinical and environmental) were sequenced by WGS over a period of 15 days (Aug. 11th to Aug. 26th)

WGS

Same PFGE pattern

Minimum Spanning Tree Analysis

Other isolates with same PFGE pattern from different locations and back in time to 2007 also had 0‐3 SNPs

= 1 SNP difference 1

Reference genome?

11/28/2017 25

Hospital X was challenging that Patient D could not have been infected with Legionella in their facility

• Any match in entire tree that is 0-20 SNPs
• Assess samples from same facility, same

geographic area

• Email (patient initials)
• NYS and/or NYC Epi, NYC PHL, Env.

Health, DOH Leadership

What, how, and who…

NTM, Neisseria meningitidis, etc…

• 8 cases of Neisseria

meningitidis serogroup B closely linked by geography and time- are these related by WGS?

• Medical Tourism cases

with Mycobacterium abscessus- can we provide WGS on cases from NYC and other states?

11/28/2017 26

We don’t know what is coming around the corner…

Continue to have Lab and Epi discussions improve with feedback and experience.

Acknowledgements

• Wadsworth Center

– Bacteriology Laboratory – Mycobacteriology Laboratory – Applied Genomics Technologies Core – Bioinformatics Core – CLIMs Group

• Epidemiology-NYS, NYC, Env. Health
• NYC PHL
• WC, DOH Leadership
• CDC, FDA, NCBI
• State Collaborators