The Human Genome at Ten

Jutta Marzillier, Ph.DLehigh University

Biological SciencesBiological Sciences

December 3rd , 2010

The Human Genome at 10 H did i h i d dHow did it start, how was it done, and

what is its Impact on Science and Medicine?

Jutta Marzillier, Ph.DLehigh University

Biological Sciencesg

December 3rd , 2010

10 years ago in June 2000 at the White House:First rough draft of the “book of life” announced

Clinton:Clinton:Completion of the Genome Project would“…revolutionize the diagnosis, preventionand treatment of most, if not all, human

Craig Venter Bill Clinton Francis Collins

diseases……”

Collins:Collins:…”personalized medicine is likely to emergeby the year 2010…..”

10 years ago at the White House:First rough draft of the “book of life” announced

2010 Th t 102010: The next 10 years

Craig Venter Bill Clinton Francis Collins

YearYearofof

2010 21 Jan 2010 11 Feb 2010 18 Feb 2010

……the genomic Era………..

4 Mar 20101 Apr 2010

29 Apr 2010 7 May 2010

O liOutline

•Rationale for Genome Sequencingq g

•Method of DNA sequencing

•DNA sequencing technologies

•Major insights and achievements

•What’s next?

Rationale: Why is the Knowledge about the Human Genome interesting?

http://www.genomenewsnetwork.org/articles/06_00/sequence_primer.shtml

The human body has about 100 trillion cells.Each cell harbors the same genetic information in its nucleus in form of

http://www.pharmainfo.net/files/images/stories/article_images/

Each cell harbors the same genetic information in its nucleus in form of DNA containing chromosomes.Depending on cellular, developmental, and functional stage of a cell only a subset of genes is expressed. g pModifications in the genetic information may result in disease.

Why is Genome Sequencing Important?

• To obtain a ‘blueprint’ – DNA directs all the instructions needed for cell development and function

• To study gene expression in a specific tissue, organ or tumorTo study gene expression in a specific tissue, organ or tumor• To study how humans relate to other organisms • DNA underlies almost every aspect of human health, both, in

function and dis-function•To find correlations how genome information relates to development

of cancer, susceptibility to certain diseases (personalized medicine) and drug metabolism (pharmacogenomics)

What is DNA?DNA constitutes the heritablegenetic information that forms the basis for the developmentalbasis for the developmental programs of all living organisms.

A genome is an organism’s complete set of DNA

Watson & Crick, 1953

complete set of DNA

The DNA is made up of fourbuilding blocks called nucleotides.g

DNA sequence is the particular side-by-side arrangement of basesalong a DNA strandalong a DNA strand.

DNA sequencing is a biochemical method to determine the sequence of the nucleotide bases that make up the DNA.

Method of DNA sequencing:Principle of DNA Synthesis

Arthur Kornberg demonstrated DNA replication in aArthur Kornberg demonstrated DNA replication in acell-free (in vitro) bacterial extract (Nobel prize, 1959)

Di d DNA l (P l1) t f ilit t DNA th i-Discovered DNA polymerase (Pol1) to facilitate DNA synthesis- Unraveled the mechanism of DNA synthesis

- building blocks (Adenine, Cytosine, Guanine, Thymine)g ( , y , , y )- a single DNA strand serves as a template- can only extend a pre-existing chain (primer)- a free 3’ Hydroxyl end is required

Watson et al., MGB, 2008

Reaction Mechanism for DNA synthesisReaction Mechanism for DNA synthesis

The 3’ hydroxyl group of the primer attacks the -phosphoryl group of the incoming nucleotide thereby forming a phosphodiester bond (SN2 reaction).nucleotide thereby forming a phosphodiester bond (SN2 reaction).

W t t l MBG 2008Watson et al., MBG, 2008

Did S iDideoxy Sequencing according to Sanger

Both nucleotide typescan be incorporated into

Frederick SangerNobel Prize (1980)

ca be co po ated togrowing DNA chain.

Presence of dideoxy-cytosine in growing chain blocks further additionf i i l tidof incoming nucleotides.

Watson et al., MBG, 2008

Dideoxy Method of Sequencing (Sanger, 1975)

‘Normal’ DNA synthesis:DNA strand as templatePrimerDeoxynucleotidesPolymerase enzymeUse several cycles to amplify

DNA sequencingDNA synthesis is carried out in

the presence of limitingthe presence of limiting amounts of dideoxyribonucleosidetriphosphates that results in chain termination

• Through chain termination• Through chain termination fragments of distinct sizes are generated that can be separated by gel electrophoresisOriginal method sed radio GCTACCTGCACCA

GCTACCTGCACCAGA

• Original method used radio-labeled primers or dideoxynucleotides GCTA

GCTACCTGCA

GCTACCTGCACCA

Dideoxy Method of Sequencing

Original method sedOriginal method sed radio labeled primerOriginal method usedOriginal method used radio-labeled primer or dideoxynucleotidesdideoxynucleotides

This method required four separate DNA This method required four separate DNA synthesis reactions to be separated by synthesis reactions to be separated by electrophoresis in four parallel laneselectrophoresis in four parallel laneselectrophoresis in four parallel lanes. electrophoresis in four parallel lanes. The gel needs to be dried, exposed to film, The gel needs to be dried, exposed to film, developed and manually read.developed and manually read.

Approx. Approx. 150150-- 300 300 bases read lengthbases read length

Summary – Sequencing Method Established

Improvements-Use of fluorescently labeled dideoxy-

- Need of four reactions in parallel - Heat labile polymerase- Use of radioactivity- Low resolution on gels

y ynucleotides

-One-lane electrophoresis-Introduction of capillary electrophoresis to

increase resolution (up to 1,000 ntes)g- Approx. 150 nucleotides read length- time consuming

( p )-Use of heat stable polymerase (Taq)- Automation

Fluorescently labeled ddNTPs used for uo esce t y abe ed dd s used osequencing reaction allow one lane analysis

ddNTPsddNTPs used for used for automated sequencing automated sequencing are labeled with are labeled with differentdifferent fluorescent fluorescent dyes representing each dyes representing each of the 4 basesof the 4 bases

dnaseq.genomics.com.cn/dnaseq/en/tech/images

Allows separation of DNA productsin same lane of gel

Capillary Gel Electrophoresis IncreasesCapillary Gel Electrophoresis Increases Resolution

Capillary gel electrophoresis:

Sequence ladder by

Capillary gel electrophoresis:Samples passing a detection window are excited by laser and emitted fluorescence is read by CCD camera. Fluorescent signals are converted intoSequence ladder by

radioactive sequencing compared to fluorescent peaks

Fluorescent signals are converted intobasecalls.

AdvantagegHigh resolutionRead length up to 1,000 nucleotides

ABI 310 Genetic AnalyzerOne-capillary Instrumentp y

Capillary and electrode: Negatively charged DNA migrates towards anode

DNA fragments labeled with different fluorescent dyes migrate according to their size past a laser

Sequencing Electropherogram•• Laser excites dyes, causing them to emit light at longer Laser excites dyes, causing them to emit light at longer

wavelengthswavelengths•• Emitted light is collected by a CCD cameraEmitted light is collected by a CCD camera•• Software converts pattern of emissions into colored peaksSoftware converts pattern of emissions into colored peaks

www.contexo.info/DNA_Basics/dna_sequencing.htmPlot of colors detected in sequencing sample scanned from smallest to largest fragment

Finishing Draft Sequences using Assembly g g ySoftware

Primary sequencing reads:

Finishing Draft Sequences using Assembly g g ySoftware

Primary sequencing reads:

Align sequences for homologiesAlign sequences for homologies

Green: 3-fold coverage

Yellow: 2-fold coverage

Fragment assembly to contig Goal for Genome Sequencing:Sanger: 8-fold coverage454: 30 fold coverage

Sequencing Technologies:Lif S i Th i T h th i C l b tiLife Sciences Thrive Trough their Co-laboration

with Engineering and BioInformatics

BiologistBioInformatics

ww

.itb.

cnr.i

t

Engineers

ww

How is Genome Sequencing Done?How is Genome Sequencing Done?

Clone by Clone Shotgun sequencingClone by Clone Shotgun sequencing

Break genome into random fragments,

fsequence each of the fragments andassemble fragments based on sequence overlaps

Create a crude physical mapof the whole genome by restrictionmapping before sequencing

on sequence overlaps

Break the genome into overlappingfragments and insert them into BACsand transfect into E.coli

Break these inserts in even smallerfragments and sequence

High-Throughput Whole GenomeHigh-Throughput Whole Genome Sequencing

Analysis of 384 sequencing reactions in parallel

George Church, Scientific American, January 2006, pp47-54

JGI Sequencing Facility(Joint Genome Institute, US Department of Energy)

Assume more than 20 384-capillarysequencers running simultaneouslyapprox. 700 bp per capillary runpp p p p yapprox. 3 hours per run

Approx 40 Million bases per dayApprox. 40 Million bases per dayin one facility

What Challenge Was it to gSequence the Human Genome?It took more than 6 years to determine the complete base sequence of the E coli genomecomplete base sequence of the E. coli genome (4.6 Mb). The Human Genome has 3200 Mb.

At this rate it would have taken a single lab more gthan 4,000 years to sequence the entire human genome

The average fragment read length is aboutThe average fragment read length is about500 bases.

It would take a minimum of six million (3

International Co-laborationeffort started

billion/500) to sequence the human genome(no overlap, 1-fold coverage).

Develop procedural and computational

Initial sequencing and analysis of the human genome Nature 409, 860 - 921 (15 February 2001) International Human Genome Sequencing Consortium(headed by Francis Collins)

The Sequence of the Human Genome , Science, Vol 291, I 5507 1304 1351 16 F b 2001Develop procedural and computational

algorithms and efficient database managementIssue 5507, 1304-1351 , 16 February 2001Venter et al. (Celera Genomics)

In 2001 the First Draft of the Human Genomeis Published

How many nucleotide pairs doesthe Human Genome have?

How many genes does it take

3 x 106 or 3 x 109 ?

to make a human?

Roundworm: 6,000Fruitfly: 14,000

How different are genomes from one individual to another?

y

0.1% or 1% or 10 % ?

What did we learn?What did we learn?

Less than 2% of the Human Genome codesfor proteinfor proteinThe human genome encodes for approx. 21,000 protein-coding genesThe human genome sequence is almost exactlythe same (99 9%) in all peoplethe same (99.9%) in all peopleNon-coding DNA segments may have regulatoryfunctions on gene expressionThe genome size does not correlate withthe number of estimated genesthe number of estimated genes

Organism Genome Size Est. Gene #E. coli 4.6 Mb 4,400

Yeast 12.1 Mb 6,200

Humans have only about twice thenumber of genes as a fruit fly andbarely more genes than a worm!

Roundworm 97 Mb 19,700

Fruit Fly 180 Mb 13,600

Rice 389 Mb 37,500

Human 3200 Mb 25,000,

www.ornl.gov/hgmis/publicat/primer

Who’s sequence was used for the ‘Human Genome Project’Genome Project

• Different sources of DNA were used for original sequencing• Celera: 5 individuals; HGSC: ‘many’y• The term ‘genome’ is used as a reference to describe a

composite genome• The many small regions of DNA that vary among

individuals are called polymorphisms:individuals are called polymorphisms:– Mostly single nucleotide polymorphisms (SNPs)– Insertions/ deletions (indels)– Copy number variation– Inversions

http://scienceblogs.com/digitalbio/2007/09/genetic_variation_i_what_is_a.php#more

SNPs: the human genome has at least 20 million SNPsmost of these SNPs contribute to human variation

f th i fl d l t f di tibilitsome of them may influence development of diseases, susceptibilityto certain drugs, toxins, infectious agents

Development Sequencing Time Linep q g

1953:Discovery of theDNA double helixby Watson & Crick

2005

2006

The Race for the $1 000 GenomeThe Race for the $1,000 Genome

Human Genome Project (2001 initial draft):Human Genome Project (2001, initial draft):> $ 3 billion (includes development of technology)“raw” expenses estimated at $300 million

Rhesus macaque (2006)$ 22 million

By end of 2007:$ 1-2 million for full mammalian genome sequence

Wanted: “!!!!!! The $ 1,000 Genome !!!!!!!!!”- low cost- high-throughput - high accuracyg y

454 Pyrosequencing Method454 – Pyrosequencing Method

Flowgram:Flowgram:

Add one kind of dNTP per cycle. Sulfurylase converts PPi to ATP in ypresence of APSLuciferin conversion results in light emission

Intensity of light emission is directlyProportional to amount of releasedPPi l lPPi moleculesSequencing is recorded ‘live’. Read length up to 400 basesProblem: homopolymer detection

Genome Sequencing Milestones

May 2007: James Watson’s genome deposited454 technology (pyrosequencing)2 month, $ 1-2 million, $

Sept 2007: Craig Venter’s genome depositedSanger technology, 4 years, 70 million

*Diploid genomes

*Approx. 3.5% of Watson’s genome could not be matched to the reference genome

Appliedbiosystems.com

pp g g

*Venter’s genome had 4.1 million DNA variants comprising 12.3 Mb- 3.2 million SNPs- non-SNP variants

Personal Genome Project (PGP): initiated by George Church (Harvard) seeking for volunteers to share their DNA sequence and medical records

Next Generation Sequencing (NGS) Focuses onMiniaturization and Parallelization

Sanger Cyclic Array Sequencing

Generate dense planar arrayf DNA f tof DNA features

Apply cycles of enzymatic driven biochemistryy

Imaging-based data collection

Sequence Mate Pairs Provide an AdditionalSequence Mate-Pairs Provide an Additional Layer of Positional Information

DNA fragmentation innebuilizer, end polishing,size selection by DNA gel

l t h ielectrophoresisAdapter ligation and sequencing

-

1000

500

200

Sequence mate-pairs separated by a known distanceAssemble contigs into larger scaffolds using information about

00

+

sequence similarity and distance between mate-pairs

Next generation sequencing – 454 method(P i )(Pyrosequencing)

-Shear DNA into 300 to 400 nte fragments

-Ligate to heterobifunctional adapters

-Capture DNA strands on beads and amplifiyby emulsion PCRby emulsion PCR

-Place beads with amplified DNA fragmentsinto picotiter plate (1.7 Mio wells)

- Perform pyrosequencing reaction

44 m

Solexa/ Illumina SequencingTechnology

reversibleSangerIllumina

Use a set of deoxynucleotides that carry

reversible

Use a set of deoxynucleotides that carry

- each a fluorescent label that can be cleaved off

- a reversibly terminating moiety at the 3’ hydroxyl position

Solexa/ Illumina SequencinggTechnology

Comparison

Sanger Sequencing:In vitro construction of library

454 Sequencing:Direct in-vitro amplificationIn vitro construction of library

Limited parallelization‘large’ volumes (microliter)Expensive

Direct in vitro amplificationMassively parallelReduction of reagent volume to pico or femto-liters per DNA feature

R l ti l h

Long reads (up to 1000 bases)

Relatively cheap

Shorter read length (300-400 bases)g ( p )High accuracy (0.1% error)

4 x 105 bases/run

g ( )Lower raw accuracy (1% error)

6 x 108 bases/run

Illumina sequencingRead length 36 – 50 bases

2 x 109 bases/run

BGI – The Sequencing FactoryBGI – The Sequencing Factory(Beijing Genome Institute)

Featured in Nature issue March 4th, 2010

Purchased 128 HiSeq2000seq encers from Ill mina in Jan ar 2010

,

sequencers from Illumina in January 2010each of which can produce 25billion base pairs of sequence a day

Speed Reading1987: ABI sequencer – 4800 basepairs a day2010: Life Technologies – 100 billion basepairs a day

C. Venter, nature, 464, pg676, 2010

All published fully sequenced human genomesi 2000since 2000

Outlook: Single Molecule Sequencing through Graphene nanoporeGraphene nanopore

•Graphene (Geim & Novoselov, Nobel Prize 2010)is a planar sheet of carbon just one atom thickis a planar sheet of carbon just one atom thick•Strongest material known, electricallyconductive

As a DNA chain passesAs a DNA chain passes through the nanopore, the nucleobases, which are the letters of the genetic code, can be identified. The nanopore in graphene is the first nanopore short enough to distinguishenough to distinguish between two closely neighboring nucleobases

Achievements:The duck billed platypus: part birdThe duck-billed platypus: part bird, part reptile, part mammal – and the genome to prove itgenome to prove it

http://news.nationalgeographic.com/news/bigphotos/4891235.html

Th Pl t i t di t t l l ti- The Platypus is our most distant mammal relative- Has retained a large overlap between two very different classifications- The platypus shares 82% of its genes with human,

d d hi kmouse, dog, opossum and chicken - Decoding the platypus genome helps to understand the origins of mammal evolution

Nature, May 7, 2008

Decoding of a Acute Myeloid Leukemia Genomeg y

* AML i f hit bl d ll hi h l t i th b d* AML is a cancer of white blood cells, which accumulate in the bone marrow andinterfere with the production of normal white and red blood cells.

* Sequenced and compared the genome of cancerous bone marrow cells withnormal skin cells from the same patient using the next generation technologies.

* F d t t l f 10 i t ti i th ll* Found a total of 10 genes carrying mutations in the cancerous cells. * Two of those (FLT3 and NPM1) had already been implicated in the process of

progression to AML.

Proof of concept that whole genome sequencingf t ll i f l t t t

Ley et al., nature 456:66, 2008

of tumor cells is a very useful strategy to uncovergenes implicated in the disease.

Ley et al., nature 456:66, 2008

Twin study surveys genome for cause of Multiple Sclerosis

MS causes the body’s own immune cells to attack the myelin sheath around nerve cells

MS has a genetic component

No clear genetic reason found to explain whyone twin developed MS while the other did not

Both twins carry genetic variants that are linkedto a higher risk getting MS

Baranzini et al

to a higher risk getting MS

But those genetic factors seem to have been insufficient to cause disease of their ownBaranzini et al.

April 29th, 2010 Environmental triggers, epigenetics?

Common Gene Variant Strategygy

Assumption:pTo predict human diseases orsusceptibility common DNA variations (~5% frequency)would be at a faultwould be at a fault.

Many SNP alleles have been uncoveredrelated to specific diseases. however, these variants have onlyaccounted for a small fractionof disease risk.

“A gene is vexed with multiple layersof complexity” (Joseph Nadeau)

Hall. Revolution Postponed. Scientific American, Oct. 2010, pg 60-67

In Contrast:Look at Rare Variants and Beyond the Genes

Other mechanisms ofgene expression regulation:

Inhibitory RNA (RNAi)

Chemical DNA modification, e.g. methyl groups

Lessons learned• Data sharing

• 2000: 4 eukaryotic genomes• 2010: 250 eukaryotic genomes many hundreds of human genomes2010: 250 eukaryotic genomes, many hundreds of human genomes

• Fewer protein-coding genes than thought

S f• Studying of genetic variation and evolutionary origins

• Cost of sequencing has fallen 100,000 fold; high trough-put sequencing

• HapMap project: Catalog more than 20 million SNPs

• Cancer genome atlas to map genomic changes observed in every major type of hhuman cancer

• Obtaining a genome ‘blueprint’ may be sufficient to explain the susceptibility to disease ,, but not the likelyhood to develop the disease. p y ,, y p

• The combination of m a n y factors needs to be taken into account

Ethic Aspects of Genome Sequencing

-How can patient privacy be protected?-Will insurance companies and employersp p y

use genetic information to screen outthose at high risk for disease?

-Genetic engineering of bioterrorism agents

nhs.needham.k12.ma.us

What’s next?What s next?

Conference Topics:• Genetic basis of common diseases• Intricacies of gene regulation• Role of non coding portions of the genome• Role of non-coding portions of the genome• Medical advances• Technological and ethical challenges of human genomics• Personalized genomics• Aspects of variation in the human genome