Bio conference live 2013

What to use for targeted re-sequencing – microarrays or Next

Generation Sequencing?

Agnieszka M. Lichanska TPCaruso & associates, Durham, NC

Choices for gene-c tes-ng

Content 1.  Learning objectives 2.  What is re-sequencing?

•  Why do we use it? •  When do we use it? •  What are the types of re-sequencing methods?

3.  Technologies used for re-sequencing – technical comparison •  Workflows •  Comparison of performance •  Data analysis

4.  Establishment of a new assay in the laboratory •  Requirements •  Validation •  Reference materials •  Reporting and interpretation •  Regulatory oversight

5.  Summary 6.  Questions

TPCaruso & associates BioConference Live Genetics and Genomics 2013

Learning Objectives

1.  Understand how targeted re-sequencing is applied to genetic screening

2.  Appreciate the complexity of validation protocols

for re-sequencing assays.

BioConference Live Genetics and Genomics 2013 TPCaruso & associates

Re-sequencing

•  It is sequencing of a part or whole genome of an individual in order to detect sequence differences between the individual and the standard genome of the species


Re-sequencing

•  When to use it? o  Carrier screening o  Prenatal screening o  Newborn screening o  Undiagnosed childhood diseases/disorders o  Risk assessment for complex diseases

•  Types of methods used: o  Targeted re-sequencing panels o  Exome sequencing o  Whole genome sequencing

•  Applications: o  Genotyping o  Variation screening


How to perform targeted re-sequencing?


Exon Regulatory region Amplicon Flanking region Buffer region Primer

Target

Amplicons

Probe Gene X, area of interest

DNA

Beads

Gene X, area of interest

DNA

Array Probe

In solu(on capture On array capture

PCR amplifica(on

MPX PCR & LR_PCR RainDance ePCR Fluidigm Haloplex Ion Torrent Illumina

Changing face of genetic testing


Complexity

Num

ber o

f gen

es

1

>10,000

10

100

1000

Low Medium High

Sanger

Arrays

NGS

Assays available for targeted re-sequencing


Technique Assays

Sanger Sequencing ABI Resequencing Sets (RSSs)

Array-‐based high-‐throughput re-‐sequencing strategies

Pan-‐Ethnic Carrier Screening Panel *Noonan Syndrome *Cardiac myopathy *Charcot-‐Marie Tooth

Next GeneraUon Sequencing Carrier screening panel, incl. CFTR panel Noonan syndrome panel Cardiac diseases panel Pharmacogenomics panel Intellectual disability AuUsUc disease spectrum panel Pulmonary diseases panels NeurodegeneraUve disorders MulUple cancer panels and more

* The assay is no longer used, transiUoned to NGS

TECHNICAL COMPARISON


Re-sequencing microarrays

•  Formats and total number of bases that can be sequenced o Cartridge:

o 49 format – 317 kb o 100 format – 117 kb o 169 format – 47 kb

o Strips of 4 chips o  96-well plates of chips (High Throughput

Assay or HTA)


TPCaruso & associates

1234!GAGACGGATTCTCCGCCGAGCTGTCCGATACG!CTCTGCCTAAGAGGCGGCTCGACAGGCTATGC!

Reference sequence

Patient’s Sample sequence

GAGACGGATTCTCCACCGAGCTGTCCGATACG!------------GGTG----------------!

GAGACGGATTCTTCGCCGAGCTGTCCG!GAGACGGATTCTGCGCCGAGCTGTCCG!GAGACGGATTCTCCGCCGAGCTGTCCG!------------GGTG-----------!GAGACGGATTCTACGCCGAGCTGTCCG!

AGACGGATTCTCTGCCGAGCTGTCCGA!AGACGGATTCTCGGCCGAGCTGTCCGA!AGACGGATTCTCCGCCGAGCTGTCCGA!-----------GGTG------------!AGACGGATTCTCAGCCGAGCTGTCCGA!

GACGGATTCTCCTCCGAGCTGTCCGAT!GACGGATTCTCCGCCGAGCTGTCCGAT!GACGGATTCTCCCCCGAGCTGTCCGAT!GACGGATTCTCCACCGAGCTGTCCGAT!----------GGTG-------------!

ACGGATTCTCCGTCGAGCTGTCCGATA!ACGGATTCTCCGGCGAGCTGTCCGATA!ACGGATTCTCCGCCGAGCTGTCCGATA!---------GGTG--------------!ACGGATTCTCGCACGAGCTGTCCGATA!

Base # 1

Base # 2

Base # 3

Base # 4

Re-sequencing microarrays difficult sequences


GAGACGGATTCTCCGCCGAGCTGTCCGATACG!CTCTGCCTAAGAGGCGGCTCGACAGGCTATGC!

Reference sequence

GAGACGGATTCTCC---GAGCTGTCCGATACG!------------GG---CTC------------!

GAGACGGATTCTCCGTTTCCGAGCTGTCCGATACG!------------GGCAAAGGCTC------------!

GAGACGGATTCTCCGCCGAGCTGTCCGATACG!CTCTGCCTAAGAGGCGGCTCGACAGGCTATGC!

Reference sequence

GAGACGGATTCTCCTCCTCCGGGGGGTCCGATACG!CTCTGCCTAAGAGGAGGAGGCCCCCCAGGCTATGC!

GAGACGGATTCTCCTCCTCCGGGGGGGGTCCGATACG!CTCTGCCTAAGAGGAGGCAGCCCCCCCCAGGCTATGC!

Repeats

Single base changes in homopolymer

stretches

Insertions

Deletions

Large deletions Exon 1 Exon 2 Exon 3

Next Generation Sequencing


36 -‐ 250bp Up to 200bp Up to 1000bp 50 – 75bp Up to 20,000bp

Fluorescence H+ ion detecUon Luminescence Fluorescence Fluorescence

Bridge amplificaUon

ePCR ePCR ePCR No amplificaUon

Homopolymer errors

Homopolymer errors

slow Low yield at high accuracy

Signal/Noise RaUo

Sequence

Sequencing workflows – Microarray vs NGS


DNA FragmentaUon

Sequencing

Data Analysis ReporUng

gDNA template

MPX PCR amplificaUon

Pooling, purificaUon, FragmentaUon, labeling

HybridizaUon

Wash & stain Scanning

Targeted enrichment:* Capture or amplificaUon

Tagging of the fragments*

* Order of steps varies on a protocol used

Microarray NGS

NGS output example


CTTACAGATATGTGTTGAGACGGATTCTCCGCCGAGCTGTCCGATACGCCCATGAAAAGCT!

AARS c.986G>A CTTACAGATATCTGTTCA CTTACAGATATCTGTTGAGA CTTACAGATATCTGTTGAGACGGAT CTTACAGATATCTGTTGAGACGGAT CTTACAGAAATCTGTTGAGACGGAT CTTACAGATATCTGTTGAGACGGATTCT CTTACAGATATCTGTTGAGACGGATTCTCC CAGATATCTGTTGAGACGGATTCTCCACCG CAGATATCTGTTGAGACGGATTCTCCACCG GATATCTGTTGAGACGGATTCTCCACCGAG

ATAT-TGTTGAGACGGATTCTCCACCGAGC GATATCTGTTGAGACGGATTCTCCACCGAG

TGTTGAGACGGATTCTCCACCGAGCTGTCC TGTTGAGACGGATTCTCCACCGAGCTGTCC

GAGACGGATTCTCCACCGAGCTGTCCGATA GAGACGGATTCTCCACCGAGCTGTCCGATA

GAGACGGATTCTCCACCGAGCTGTCCGATA GATTCTCCACCGAGCTGTCCGATACGCCC GATTCTCCACCGAGCTGTCCGATAGGCCC TCTCCACCGAGCTGTCCGATACGCCCATG CTCCACCGAGCTGTCCGATACGCCCATGA CTCCACCGAGCTGTCCGATACGCCCATGA TCCACCGAGCTGTCCGATACGCCCATGAA TCCACCGAGCTGTGCGATACGCCCATGAA

REFERENCE SEQUENCE Coverage

COMPARISON OF MICROARRAY-BASED ASSAYS AND NGS-BASED ASSAYS


Pros and cons of each technology

Pros and cons of re-sequencing microarrays

Pros •  Quick turn around time –

24-48 hrs •  Simple experimental

protocol •  Once optimized rapid

bioinformatic analysis •  Reports sequence at each

tested locus unlike SNP arrays

Cons •  Insertions and deletions are a

problem •  Requires MPX PCR

amplification for targets •  Noise is not consistent across

the array •  Costly modifications to

content •  Mutations can only be

detected in tiled regions •  Problematic templates:

–  Partly degraded DNA –  Whole genome amplified DNA


Pros and cons of NGS

Pros •  Detection of insertions

and deletions •  Detection of novel

mutations •  Better reproducibility •  Less incidental findings •  Continually decreasing

pricing

Cons •  Not full coverage in

some areas •  More complex sample

preparation •  Development of

bioinformatic analysis is complex

•  Sequence length can be an issue

•  Risk of over-interpretation of results


SUMMARY COMPARISON OF MICROARRAY-BASED ASSAYS AND NGS-BASED ASSAYS


Feature Microarray NGS

Development time Same as NGS – 50% Shorter – 50%

Turn around time In majority cases shorter

Reproducibility - 100%

Data Analysis Data complexity Pipeline/report development More accidental findings More pathogenic mutations Higher confidence in data

No difference – 50% Similar

50% Similar No Yes Yes

Data Quality: Less “N” calls Substitutions detection Indels detection

- Same Not adequate

100% Same or slightly better Better

Cost effectiveness Varies Varies

Validation Similar or slightly slower Similar or slightly quicker

Data Analysis/Reporting


Sequence

1.  QC •  Experiment •  Obtained sequences •  Remove poor sequences

2.  Align to the reference genome 3.  Filter out known SNPs (non-

pathogenic) 4.  Identify mutations 5.  Create a report

•  Sample name •  Test name •  Laboratory QC metric •  List of mutations, zygosity •  Frequency of alleles

SeqNext: Evaluation of data analysis software for next generation sequencing in a clinical laboratory

Jakub Patrick Sram, Harry Gao, Dongqing Gu, Wengang Chen, Juan-Sebastian Saldivar, Lawrence Weiss

City of Hope, Duarte, CA 91010Abstract

Next generation sequencing is revolutionizing clinicalmolecular diagnostic testing. Many CLIA laboratoriesare adapting this innovative sequencing technology todevelop new clinical tests to answer unmet needs, suchas sequencing large gene panels in a cost efficientmanner. As next generation sequencing replacescapillary sequencing in the clinical space, the cost perbase becomes cheaper and cheaper, but the dataanalysis bottleneck still remains. Present medicalgeneticists and technologists need user-friendlysoftware that will help them manage data, analyze forand find unknown mutations and report resultsproduced by the next-gen sequencers. Unfortunately,the number of software products on the current marketthat can perform such a task is extremely limited.

Our objective was to thoroughly evaluate, analyze andvalidate the new software, SeqNext, developed by JSIMedical Systems. We used the Illumina GA IIx tosequence 12 p53 patient samples with 83 knownalterations that were previously sequenced by an ABI3730 capillary sequencer. The 83 mutations in the p53patient samples included: 48 bp, 16 bp, 2bp and 1bpdeletions; a 7bp insertion; indels (ins 5bp del 1 bp) andsingle base substitutions. SeqNext correctly detected100% of the 83 mutations in the p53 samples with oneoptimized analysis setting.

Complex mutations such as two overlapping deletionsor insertions are exceptionally difficult to detect bycapillary sequencing. SeqNext detected all sequencevariants in a mixture of three p53 samples, includingtwo overlapping insertions (7bp and 4bp). In addition,we evaluated competitive software commerciallyavailable on the market but we were not able to detectthe more complex mutations. We found SeqNext to bea user-friendly software with the potential to significantlyreduce data analysis time without sacrificing accuracyin detecting unknown mutations.

Fig 1. Example of a complex hetorozygousmutation sequenced by CE where a T base is deleted and CCCAA bases are inserted

Conclusions

430

Work flow

• Long PCR from exon 2-10 of p53• DNA shearing • Adding adapter and barcodes• Mix samples at about equal ratio• GAIIx sequencing• Sort data by barcode• Mutation analysis with SeqNext software• Compare with capillary sequence data

del.T/ins.CCCAA (capillary)

Limitations For Capillary Sequencing

• High background• Difficult to analyze heterozygous deletions,

insertions and indels• Much more difficult to analyze complex

mutations– two overlapping deletions or insertions

Ins7bp & Ins4bp

48bp & 16bp overlapping deletions

raw data view: 48bp deletion

T>G point mutation

T>G point mutationraw data

Low level mosaic mutation

Sample Reportraw data view: 16bp deletion

Fig. 2. A mixture of three p53 samples was sequenced by GAIIX and SeqNext detected correctly two heterozygous insertions of 7 bpand 4 bp, CGGTTTC and CCCA, respectively.

• SeqNext correctly detected all complex mutations under one setting.

•We found SeqNext to be a user-friendly software

•Using GAIIx sequencing and SeqNext, low level mosaic mutations that otherwise could not be detected by Sanger sequencing can now be detected

Fig. 3. In a mixture of three p53 samples, SeqNextwas able to detect two large hetorozygousdeletions of 48bp and 16bp at the same sequence position.

Fig. 4. 48bp deletion visualized by aligning reads to the p53 reference sequence.

Fig. 5. 16bp deletion visualized by aligning reads to the p53 reference sequence.

Fig. 6. Heterozygous T>G mutation visualized by SeqNext

Fig. 7 . Heterozygous T>G mutation visualized by SeqNext by looking at the actual reads

Fig.9. Low level mosaic mutations can be detected by GAIIx sequencing and SeqNext software.

Fig.10. Actual reads visualized by SeqNext for the mosaic C>A change

Fig.8. SeqNext can print a clinically relevant report

Low level mosaic mutation raw data

WHICH ONE TO CHOOSE THEN?

Choice is based on a number of factors: 1.  Required throughput 2.  Nature of mutations 3.  Complexity of the assay to be performed 4.  Availability of equipment and expertise 5.  Time required for implementation of the assay


ESTABLISHING TARGETED RE-SEQUENCING ASSAY IN THE

LABORATORY

1.  Reasons for introduction 2.  Choice of platform 3.  In-house development or off the shelf product 4.  Validation – technical and clinical 5.  Implementation


1. Reasons for introducing a new assay

•  Streamline the tests available – instead of 10s of tests just one

•  Provide more comprehensive testing •  Meet the clinical needs of the customers •  Better performance, more accurate, more specific •  Competition already using gene panels •  Improve turn around time


FASTER – BETTER - CHEAPER


Optimization of a test

VALIDATION ASSAY

PERFORMANCE QC

PROFICIENCY TESTING DEVELOPMENT

Platform Assay/Test IT/pipeline

Patient testing

IMPLEMENTATION

Daily or

when test is run

Periodical •  Assay content •  OpUmizaUon of capture

or amplificaUon •  OpUmizaUon of

sequencing process •  Technical validaUon •  Development of analysis

Platform and assay source

2. Choice of platform NGS vs microarrays Type of NGS instrument Enrichment method

3. In-house development, off the shelf product or collaboration with another company Regulatory requirements Time and costs of assay development Time and costs of assay validation Availability of assay analytical performance documentation


4. Validation of a re-sequencing assay

•  Definition of performance specifications –  Analytical sensitivity – likelihood of detecting

sequence variation if present –  Analytical specificity – false +ve rate –  Reportable range – genes, exons, genomic regions –  Reference range – reportable sequence variations

that can be detected –  Necessary coverage – to make accurate calls

•  Establishment of QC process for the performance specs –  Coverage, quality scores, strand bias, GC bias,

transition/transversion ratio, allelic read percentage •  Establishment of proficiency testing (PT)


Reference materials – critical part of validation and on-going QC protocols

•  gDNA (blood or cell lines) –  Blood (non-renewable) –  Cell lines (renewable) but can have genetic

aberrations –  Similar to patient’s sample

•  Synthetic –  Not representative of gDNA complexity

•  Electronic reference materials –  Reference for analysis steps –  Data files might not be exchangeable between

platforms


Example of reference materials


5. Implementation •  Regulatory oversight

–  FDA requirements – LDT vs cleared/approved tests –  CLIA regulations –  State requirements

•  Guidance –  CDC –  Clinical Lab Standards Institute –  ACMG –  AMP –  CAP

•  Proficiency testing and QC –  No formal programs exist for NGS –  QC has to include sample prep, library prep, generation of

sequences, analysis of sequences and result reporting


Implementation – Reporting guidance

•  Constitutional mutations in genes (57) listed by ACMG on a minimum list – reported independently of the test indications

•  Additional genes – might be analyzed for incidental findings (up to the lab)

•  Incidental findings – report irrespective of the patient’s age

•  Incidental findings – reported for a normal not tumor tissues


Who is going to provide counseling to the patient?

•  According to ACMG the ordering clinician or their team is responsible for providing the counseling

•  A few points on this: – Clinicians need to be familiar with NGS and its

limitations – Explain to the patients a possibility of incidental

findings – A clinical geneticist should be consulted regarding

the ordering, interpretation and communication of the complex NGS assays.


Summary

•  Targeted sequencing allows sequencing of panels of genes for particular conditions

•  Development and validation are complex – Reference genome – Reference templates – Alternative tests for sequences with no or limited

coverage •  Reporting requirements

– Combines QC of the assay and genetic data – Simple yet comprehensive


Bio conference live 2013

Health & Medicine

Transcript of Bio conference live 2013