SMRT-LeidenJun 12-14, Leiden, NL

The perfect genome is a treasure,

DNA is the key

Olga Vinnere Pettersson, PhD

Project Coordinator

National Genomics Infrastructure

Uppsala University

Outline

National Genomics Infrastructure (NGI): who we are.

Long-read Sequencing at NGI: what we do.

HMW-DNA R&D:

• Lessons learned from 4 years with PacBio

• HMW-DNA vs ”normal” DNA

• HMW-DNA extraction lab & our experience

NGI: Mission statement

From Jan 1, 2013: NGI is a national resource for massively parallell sequencing

National

resource

State-of-the-

art

infrastructure

Guidelines

and support

Our mission is to make a state-of-the-art

infrastructure for massively parallel DNA

sequencing and SNP genotyping available

to researchers all over Sweden enabling

internationally competitive research in

genomics.

To provide guidelines and support for study

design, sample collection, protocol selection and

bioinformatic analysis.

Platform organisation

NGI StockholmNGI UppsalaNGI

NGI-Uppsala

SNP&SEQ

Technology

platform

NGI-Uppsala

Uppsala

Genome

Center

NGI-Stockholm

Ann-Christine Syvänen, UU Professor in MolecularMedicine

Lars Feuk, UU Senior lecturer(Deputy)

Joakim Lundeberg, KTH Professor in Gene Technology

Ulf Gyllensten, UU Professor in Medical Molecular Genetics

Project handling at NGI

“This research infrastructure is world class and a jewel in the crown of Swedish bioscience.” (Swedish Research Council)

Long-Read technologies in NGI service

0 20 40 60 80 100 120 140 160 180

Iso-seq

Base Modifications

Sequence capture

De novo small genomes

De novo large genomes

Long amplicons

Short amplicons

Number of PacBio projects, 2013-2017

EVERY PROJECT IS UNIQUE

SMRT sequencing, lessons learned: DNA quality requirements

For Long Reads one needs to have long and pure DNA

Same yeast, different DNAFOCUS: chemical purity

Focus on chemical purity: sorting out the culprits

"DNA chemical structure," by Madeleine Price Ball

1. Carry-over from host cells- you name it…

2. Carry-over of extraction chemicals- Phenol- Guanidinium- Ethanol- EDTA

Most have dual action: - Enzyme inhibition- DNA-binding

Focus on chemical purity: why so important?

Organism type, de novo application,

60x2 Gb / SMRT 4 Gb / SMRT 7 Gb / SMRT

Bacterium (3.2 Mb) 2-4 strains 6-8 strains 10-12 strains

Insect (300 Mb) N cells: 9Price: 10 kEUR

N cells: 5Price: 6 kEUR

N cells: 3Price: 4 kEUR

Bird (1.2 Gb) N cells: 43Price: 47 kEUR

N cells: 22Price: 25 kEUR

N cells: 12Price: 14 kEUR

Mammal (3.2 Gb) N cells: 96Price: 105 kEUR

N cells: 48Price: 53 kEUR

N cells: 27Price: 30 kEUR

Lower chemical purity = worse loading and shorter reads,

Ergo higher sequencing and analysis costs.

Focus on structural integrity: learning how to work with HMW-DNA

Working with 25+ kb libraries is becoming a standard

Do not have a Fragment Analyser of FemtoPulse yet? Consider it.

Hydroshear speed code 3Size by gel: 6 Kb

Hydroshear speed code 15Size by gel: 20 kb

25+ kb library – an example of a good DNA sample

Shearing QCInput QC

Library QCBefore size-selection

Library QCAfter size-selection

A journey to HMW-DNA world

HMW-DNA behaves very differently from LMW-DNA

High viscosity:

1. Bad hydration -> topological issues-> getting HMW-DNA in solution without any concentration gradient is an issue

2. Presence of contaminants

DNA-spa: let the molecules relax.- 3-7 days at RT or +4°C- Gentle agitation- Playing with ionic strength

FemtoPulse measurements: several replicates (and dilution series)

Getting ”correct” readings of HMW-DNA

LMW-DNA HMW-DNAHMW-DNA Read

N50, kb

E.coli in situ - fresh 13.3

E.coli in situ - relaxed 22.3

Real life examples

Bird 1

CONTIGS Primary Associated Unzipped

Primary

Unzipped

haplotigs

# contigs 2996 2136 1481 6349

# >50Kb 457 586 354 3193

Largest 52,5Mb 0,18Mb 52,6Mb 3,11Mb

N50 11,4Mb 0,05Mb 12,1Mb 0,42Mb

Total 1,12Gb 97,8Mb 1,07Gb 1,01Gb

Bird 2

SCAFFOLDS Unphased

# scaffolds 38231

# >50Kb 201

Largest 2,01 Mb

N50 0,21 Mb

Total 1,21Gb

20 kb library – perfect DNA60x

10 kb library – short, dirty& little DNA36X

User insisted to proceeddespite warning

In situ DNA extraction is a dying art (developed in late 70s – early 80s).

DNA extraction is carried out in agarose plugs -> melting -> dialysis (-> clean-up)

Allows recovery of intact Mb-size molecules, suitable for all any Long-Read technology

Other types of kits & protocols we use: MagAttract, GenomicTip & Phenol-ChISAM.

HMW-R&D at NGI: DNA lab

Nuclei extraction – worth it!

Objectives:

Testing protocols

HMW-DNA storage parameters

Preservation of HMW-fraction

Recording contaminant information

Compiling experimental data

Issue: focus on length, not chemical purity

One kit might not suit them all!

Note on commercial HMW-DNA kits

Keep an eye for:

0.5 mM EDTA -> 0.05 mM EDTAGuanidinium260/280 & 260/230

Carry-over of host cell contaminants

Kits that we know work for sure (in most cases), as on May 2018:

MagAttract, Genomic Tip, Zymogene genomic DNA clean-up and concentrator

CONCLUSIONS

- Both molecule length and chemical purity of input DNA are equally important to successful sequencing.

Spending time on optimizing DNA extractions save money for sequencing(and analysis).

- Working with HMW-DNA is a craft. It is worth to learn.

- The HMW-DNA field is becoming extremely popular.There are still many unknowns, however, the basic biochemistry is still the same.

Acknowledgements

Mai-Britt Mossbech, PhDR&D engineer, DNA extractions

Tomas Klingström, PhDFaithful young padawan

Inger JonassonBoss who allows me to play with DNA

UGC-NGI teamWho tolerates all my requests for additional DNA-QC

Sponsors and vendors