NGS in clinical diagnostics · NGS routine diagnostics at Örebro University Hospital 2013 Neisseria meningitidis Epidemiological surveillance –2 / year 2016 N. meningitidis + MRSA

NGS in clinical diagnostics

Paula MöllingMolecular Biologist, Associate professor

Dept of Laboratory Medicine, Clinical Microbiology,

Örebro University Hospital, Sweden

Content

• Routine diagnostics - Pathogens

• NGS technology - MiSeq Illumina

• Outbreak situation – Neonatal dept

• Speed up to get faster answers - Nanopore

• Sharing data nationally - Surveillance and outbreaks

NGS routine diagnostics at Örebro University Hospital

2013 Neisseria meningitidis

Epidemiological surveillance – 2 / year

2016 N. meningitidis + MRSA

Genetic typing routinely / Epidemiological surveillance – every month

2018 N. meningitidis, MRSA + N. gonorrhoeae

Genetic typing routinely / Epidemiological surveillance – every two week

2019 N. meningitidis, MRSA, N. gonorrhoeae + Clostridioides difficile

Genetic typing routinely / Epidemiological surveillance – every week

MISEQ - ILLUMINA

Library prep

(NexteraXT)

2h hands-on

6-8 h

MiSeq seq

20 min

hands-on

1.5 – 2.5 d

Analysis

Different

pipelines

few min - hours

In total 3-4 days from extracted DNA

Databases for typing of bacteria:

• Ridom SeqSphere+, http://ridom.com/seqsphere/ - MRSA, C. difficile

• 1928D, https://1928diagnostics.com/ - MRSA, C. difficile

• BIGSdb, http://pubmlst.org/software/database/bigsdb/ - N. meningitidis

• CLC Geomic workbench - N. gonorrhoeae

Assembly: Put all sequence reads together to contigs - ”puzzle”

Velvet Optimiser, CLC Genomics Workbench, SPAdes

NGS data analysis

Ridom SeqSphere+

- pipeline automatically

MRSA NGS - all new clinical MRSA + S. aureus with toxins

Analyse NGS data:

Ridom Seqsphere+

1928Diagnostics

Spa-type, SCCmec

PVL, TSST, ETA/ETB

Typing:

MLST, 7 house keeping genes

cgMLST, >1800 genes

S.aur

genome

Accepted runs:

≥ 50x coverage

≥ 97% cgMLST Targets

Neonatal outbreak 2018 at Örebro University Hospital

Neighbour Joining tree, cgMLST of 1861 genes Minimum spanning tree within the same cluster

Takes long time to answer in an outbreak situation

MinION (Nanopore)

• Library prep - Rapid barcoding kit (12 samples)

- <1 hour

• Sequencing – Flowcell on MinION and MinIT

- a few hours per sample or <24 h per 12 sample

• Data analysis

- fast5 to fastq (EPI2ME or Guppy, Nanopore)

- Demultiplexing (Barcoding, Nanopore)

- Cluster analysis (cgMLST, 1928D pipeline)

Ion Realtime Sequencing(Oxford Nanopore Technologies)

https://nanoporetech.com/how-it-works

Ultra-long reads - up to 2 Mb (10-20 Kb)

512 Gb

Sequenced by MinION

Result MinION 1928D pipline

Clustering of MinION data - 1928D

National tracing of outbreaks – share data

1928Diagnostics – “Powerful analysis of your WGS data in minutes”

RapidShare project Aim: Share NGS data in real-time to facilitate

MRSA surveillance

Three regions: Örebro, Gothenburg, Stockholm shared

data during a time-period, October 2017 until May 2018.

No epidemiological connections could be found.

•Epidemiologic typing

•Antibiotic resistance profiling

•Virulence factor profiling

Limited time period – looking into surrounding regions

Genomic Medicine Sweden (GMS)

Aim – implement high-throughput sequencing technologies and

genomics in health care to improving Swedish precision

diagnostics.

This will be run through Genomic Medicine centers at the

seven university hospitals with the aim share data and use the

same pipelines within a common national IT-infrastructure.

GMS will focus on diagnostic areas – rare hereditary diseases,

complex genetic diseases, cancer and infectious diseases.

HCP

Public

cloud

HCP - Hitachi Content Platform, national storage (>700 Pb)

GMS - Microbiology

Goal – to obtain fast and reliable typing of

microorganisms where method and results easily

can be shared between regions for surveillance

and tracing of outbreaks in real time.

Pilot project – since sequence data from

microorganisms do not include human DNA

GMS-Micro can be run as a pilot for the entire

GMS project.

Other activities: Sharing meta data and lab methods, antibiotic resistance monitoring,

fast sequencing, clinical metagenomics

GMS-Micro Pilot

• 7 regions together with PHA performs NGS of a

“Reference dataset” of 20 MRSA (outbreaks):

- MiSeq, HiSeq, NextSeq, Ion Torrent

- SeqSphere, 1928D, CLC, Bactyper, own piplines

Aim: to define outbreak related isolates

• All data will be shared and analyzed together

Aim: to test GMS IT-infrastructure, quality of the generated NGS

data from different platforms and to define a common national

data analysis for tracing outbreaks

Conclusion

Using NGS facilitate surveillance and tracing of outbreaks

especially when it is speeded up and data can be

shared between regions

Martin Sundqvist

Bo Söderquist

Amaya Lagos

Marita Ekström

Anna Fagerström

Fredrik Dyrkell

Oscar Aspelin

Dimitrios Arnellos

Kristina Lagerstedt

Susanne Staaf

Thanks to

Erika Tång Hallbäck

Liselott Svensson Stadler

Åsa Lindgren

Christian Giske

Hong Fang

Lars Engstrand

Per Sikora

Hedvig Engström Jakobsson

GMS reference group