www.BioTechniques.com105Vol. 48 | No. 2 | 2010
In the late 1990s, scientists J. Craig Venter
and Francis Collins became household
names during the epic race to sequence the
humangenome.Eventuallythatraceended
in a tie in 2000, but similar levels of fame
and publicity--not to mention millions
in potential revenue--await the company
or individual that wins sequencing's latest
race: to develop and implement a so-called
third-generation sequencing system.
The sequencing community turned its
sightstowardthepromiseofthird-generation
sequencing instruments less than five years
ago, when a number of companies started
promising single-molecule sequencing
instruments that could provide faster,
cheaper genome sequencing--instruments
that would finally enable the burgeoning
fieldofpersonalizedmedicineandtakenew
fieldsofstudysuchastranscriptomicstothe
next level.
So strong is the interest in new
sequencing technologies that the National
InstitutesofHealth(NIH)issupportingthe
race toward third-generation technologies
through their $1000 Genome Initiative,
which provides funding to companies and
individualsdevelopinginnovativesolutions
aimed at rapid, efficient DNA sequencing.
Giventhefrenziedpaceofdevelopmentand
the millions of dollars in financial support,
many suspect that it is only a matter of
time before someone crosses the finish
line, commercializing a third-generation
sequencing system that will finally give
researchers a full human genome sequence
for less than $1000.
Slow, long reads:
the starting line
Frederick Sanger was awarded part of the
1980 Nobel Prize in Chemistry for his
development of a method to sequence
DNA. His approach uses gel or capillary
electrophoresistoseparateDNAfragments
of differing lengths that have labeled
dideoxynucleotides incorporated at the
ends.Byidentifyingeachlabelednucleotide
in the resulting DNA ladder, the ordered
sequence of any DNA fragment can be
determined. After nearly 30 years, Sanger
sequencing, as it has come to be known, is
still being used by many researchers for its
simplicity and effectiveness along with its
abilitytoaccuratelydeterminethesequence
of long stretches of DNA.
A few years after Sanger won his Nobel
Prize, developers at Applied Biosystems
launched an automated DNA sequencer
based on his method, which used fluorescently
labeled dideoxynucleotides. Sanger's
sequencing-by-synthesis approach to DNA
sequencing,alongwiththenewlydeveloped
automated instruments, would serve as
the enabling technology for the Human
Genome Project. Although the method
has proved durable, reliable, and accurate,
it suffers from being slow, expensive, and
relativelylow-throughput--keyreasonswhy
Collins' government-funded researchers
required$3billionover13yearstogenerate
their draft map of the human genome.
Recent advances in the Sanger methodology
and associated computational
assembly tools, along with the availability
of the draft human genome sequence,
enabled the assembly of another human
genomesequencein2007,usingthedideoxy
approach. This time, though, the genome
wasofasingleindividual:J.CraigVenter(1).
Despite the obvious success of the Human
Genome Project and the other recent
whole-genome sequencing efforts, Sanger
sequencing is slowly becoming outdated.
Costs remain high and throughput is
Next-generation sequencing has pushed forward the boundaries of genetic research and enabled the
completion of a rapidly growing number of whole-genome sequencing projects. But the impending
arrival of third-generation DNA sequencing technology could change the landscape yet again.
Applied Biosystems' SOLiD sequencing system was commercialized
in 2008, vastly improving upon their Sanger methodbased
automated sequencing system. Courtesy of Applied
Biosystems.
Illustration credit: Tavares Jones
Special News Feature
Features
Sequencing's
new race
www.BioTechniques.com107Vol. 48 | No. 2 | 2010
Features
not high enough to support the growing
interest in personalized medicine and the
desiretouncoverthegeneticbasisofhuman
disease. Since these efforts could require
tens of thousands of individual genome
sequences, geneticists have been searching
for alternatives.
Claire Wade, a professor of veterinary
science at the University of Sydney
in Australia, recently completed the
sequencing of the horse genome using
Sanger methodology (2). "The horse was
one of the last sequencing projects to be
fully conducted using Sanger sequencing,"
says Wade. "It represents the peak of our
understanding of mammalian assembly on
that platform."
The next generation:
ready to go
As it turns out, the desire to improve upon
the Sanger approach and related instrumentation
over the last decade has led to
the development of a new generation of
sequencing platforms that are currently
used in labs around the world. But even
these newer and faster `next-generation'
platforms might merely be holding scientists
over until the arrival of third-generation
technologies.
In 2005, Roche 454 Life Sciences
becamethefirstcompanytocommercialize
a next-generation sequencing platform (3).
Their core technology maps genomes by
attaching beads to DNA fragments which
are then amplified, enriched, and loaded
onto a multi-well plate for sequencing.
Labeled nucleotides are then flowed across
the plate wells, where each well contains
a single bead. When a complementary
nucleotide crosses the template strand
and is incorporated, a chemiluminescence
reaction produces a signal that is imaged
andrecorded.Byperformingthenucleotide
flow step repeatedly with each nucleotide,
the current GS-FLX system can generate
400 million raw bases of data per 10-hour
instrumentrun,fromsequencereadlengths
in excess of 400 bases.
Illumina's Genome Analyzer (GA) IIx
system was introduced in 2006 and takes
a slightly different route. The GA generates
millionsofcloneclustersonasolidsupport,
which are then sequenced using reversible
dideoxy terminator­based sequencing
chemistry, which is akin to the Sanger
methodology.ReadlengthsusingIllumina's
system are shorter than those from Roche's
GS-FLX(averaging100bp),butthenumber
of sequence reads per run is dramatically
greater (the GA IIx generates 300 million
reads per flow cell, in comparison to the
GS-FLX's 1 million per run).
Applied Biosystems developed the
SOLiD system in 2008, which combines
elements of various approaches to sequence
clonally amplified DNA fragments linked
to beads. While the amplification and
attachment to beads are similar to the
Roche and Illumina platforms, Applied
Biosystems' platform relies on a unique
sequencing by ligation approach using
dye-labeled oligonucleotides, rather than
sequencing-by-synthesis. The method
actually provides two-base redundancy
in sequencing reads, which the company
says results in a higher accuracy. Similar
to Illumina's platform, the latest SOLiD 3
system generates large numbers of shorter
reads: the current SOLiD system can
generate more than 60 gigabases of raw
data from more than one billion sequence
tags per instrument run.
While shorter read lengths can cause
problemswithdenovogenomeassembly,the
increased number of reads per instrument
run--along with new computational tools
designed specifically for next-generation
sequencing analysis--have made it possible
to decode whole genomes rapidly on these
platforms, at a significantly reduced cost.
In August 2009, Illumina announced the
mappingofasinglehumangenomeatacost
of $48,000.
With sequencing projects that once
took more than a decade and billions
of dollars now being done in just over
2 weeks for a little less than $50,000,
genomics researchers are applying nextgeneration
instrumentation to projects
that7 years agoseemed impossible.Indeed,
the NIH-funded 1000 Genomes Project,
launched in 2008, involves labs around the
globe and seeks to create an improved map
of the variation found in human DNA by
sequencinggenomesfrom1200individuals.
Even more ambitious is the Genome 10K
project, an international effort to sequence
10,000vertebratespecies,orapproximately
one genome for every vertebrate genus.
The high throughput and digital nature of
next-generation sequencing technologies
have enabled their application within
emerging fields like transcriptomics, where
it is important to quantify numbers of
low-abundance messenger RNA molecules
and understand how they regulate protein
expression.
But in the end, according to Wade, it
might be the developing `third-generation'
sequencingmethodsthathavethepotential
to reveal even more about the composition
and dynamics of complex mammalian
genomes. "There needs to be sufficient read
length to span longer repetitive sequences,
and this represents a challenge with short
read length massively parallel methods,"
she explains.
In 2005, Roche 454 Life Sciences' sequencing-by-synthesis
instrument was the first next-generation platform to be commercialized.
Courtesy of Roche 454 Life Sciences.
Roche 454 Life Sciences' approach to sequencing relies on attaching DNA fragments to beads which are amplified and placed into
wells on a microwell plate. Nucleotides are flowed over the well and addition of a specific nucleotide to the DNA fragment on
the bead results in a chemiluminescent reaction that can be imaged, allowing for the determination of the DNA sequence of the
fragment attached to the bead. Courtesy of Roche 454 Life Sciences.
www.BioTechniques.com109Vol. 48 | No. 2 | 2010
Features
The third generation:
almost there
Cambridge, Massachusetts­based Helicos
Biosciencesissittingatthenexusofcurrent
next-generation sequencing systems
and the emergence of third-generation
sequencing. The company is the first to
developaplatformbasedonsingle-molecule
sequencing (4), an approach the company
calls True Single Molecule Sequencing
(tSMS). Capable of sequencing several
billion bases in one instrument run in real
time, the true advantage of single-molecule
approaches like Helicos' might just lie in
the ability to sequence without amplifying
template DNA, permitting accurate
quantification of specific RNA or DNA
molecules rapidly from very small starting
template amounts.
WhileHelicos'tSMSmayhavebeenthe
first approach out of the single-molecule
gate, other methods are beginning to
emerge.
"We anticipate that by 2013, the SMRT
sequencer will be available and able to
sequenceahumangenomeforafewhundred
dollars--and in a matter of minutes," says
Sejal Sheth, director of product marketing
atPacificBiosciences.Boldstatementssuch
as this are common in the world of thirdgeneration
development.
Pacific Biosciences made a splash at the
2008 Advances in Genome Biology and
Technology (AGBT) conference when
the company's chief technology officer,
Stephen Turner, showed early data on the
effectiveness of SMRT, their approach to
single-molecule sequencing. Turner's talk
at AGBT was quickly followed by two
articles describing the method in greater
detail (5,6). Pacific Biosciences' approach
works by sequencing strands of DNA on
chips containing thousands of zero-mode
waveguides(ZMWs).EachZMWfunctions
as a nanophotonic visualization chamber,
providing an incredibly small detection
volumeof20zeptoliters(10-21 liters).Atthis
volume,theactivityofasinglemoleculecan
be detected against thousands of labeled
nucleotides, which provides a window for
watching DNA polymerase as it performs
sequencing-by-synthesis. In each chamber,
a single DNA polymerase molecule is
attached to the bottom surface so that
it permanently resides in the detection
volume. Phospholinked nucleotides, each
labeledwithadifferentcoloredfluorophore,
are introduced into the reaction solution
at concentrations that promote enzyme
speed, accuracy, and processivity. As the
single molecule of DNA polymerase incorporates
complementary nucleotides, each
baseisheldwithinthedetectionvolumefor
tens of milliseconds. During this time, the
engagedfluorophoreemitsfluorescentlight
whose color corresponds to the identity of
a particular base. The polymerase then
cleaves the bond holding the fluorophore
in place and the dye diffuses out of the
detection volume. Following incorporation
and cleavage, the signal immediately
returns to baseline and the process repeats.
The result of this process could be very fast
sequencing--ontheorderof50nucleotides
per second--with longer read lengths than
is currently available with any next-generation
or Sanger-based system.
Expectations at Pacific Biosciences are
high. Sheth believes that with their current
development timeline,thecompanywillbe
the first to reach the $1000 genome. "We
firmly believe that we have the leading
third-generation single-molecule DNA
sequencing technology," he says.
Others disagree. "The technology
that will deliver a true $1000 genome
will not rely on light and optics," says
Zoe McDougall, director of communications
at Cambridge, UK­based Oxford
NanoporeTechnologies.OxfordNanopore
is advancing the use of protein nanopores
forDNAsequencinginaplatformthatuses
an electronic rather than an optical signal
to identify DNA bases (7). According to
McDougall,thisplacesOxfordNanopore's
approach in the unique position to drive
down sequencing costs.
Usingnanopores--smallopeningsinthe
lipidbilayerofacell--tosequenceDNAwas
proposed years ago, but the technology to
make the idea feasible has proved elusive.
McDougallsaysOxfordNanopore'sapproach
makesnanoporesequencingaviableoption.
Thecompanyusessiliconchipscontaininga
seriesofmicrowellswhereDNAsamplesare
introduced. The lipid bilayer that lies across
the top of the well gives a high-resistance
electrical seal across which voltage is sent
to drive a current toward the bottom of the
well.Thenanoporesaretheonlypointacross
which the current can flow. As each DNA
basepassesthroughananopore,itmustfirst
bindtoanadaptorcyclodextrin.Duringthis
binding event, each base blocks the flow of
current across the nanopore to a different
degree; the variations allow researchers to
identifywhichbasesarepassingthroughthe
pore.SimilartoPacificBiosciences,Oxford
Nanopore demonstrated with initial data
that their nanopore sequencing approach
will be able to map as many as 25 bases per
second (7) with the potential to sequence
very long fragments of DNA without inter-
ruption.
Another company is taking their thirdgeneration
offering one step further, not
The Complete Genomics sequencing method includes the creation of novel clusters of DNA fragments and pieces of known sequence
called DNA nanoballs. Courtesy of Complete Genomics.
Clifford Reid cofounded the company Complete Genomics in
2006 in an effort to develop a faster method to sequence DNA.
Courtesy of Complete Genomics.
www.BioTechniques.com111Vol. 48 | No. 2 | 2010
Features
onlybycreatinganoveltechnologyplatform,
but offering it as a sequencing service. In
2006, Clifford Reid, Radoje Drmanac, and
JohnCursoncametogethertodevelopanew
approach to how DNA is sequenced and
how researchers obtain their sequences of
interest.Forthenext3years,theircompany,
CompleteGenomics,workedondeveloping
their proprietary sequencing technology. In
2009,CompleteGenomicsannouncedtheir
intentiontoofferwhole-genomesequencing
services to researchers, and in February of
that same year, the company released their
first complete human genome sequence (8).
TheCompleteGenomicsmodeldiffersfrom
othernext-generationsequencingcompanies
in that they do not sell their instruments;
instead, the company sequences DNA
samples that customers provide, and then
reports back the results.
Complete Genomics published
schematics of their sequencing system
in November 2009 (9). The approach is
to create 500-bp libraries with genomic
DNA fragments of known sequence interspersed
at regular intervals. These are then
amplified in solution, in a single reaction
chamber, which enables higher density and
lower reagent usage. These resulting DNA
nanoballs (DNBs) consist of more than
200 copies of the head-to-tail concatemers
in a ball-like configuration. The DNBs are
then transferred onto patterned silicon
substrates with arrays of spots that are
activated to capture and hold the DNBs in
place, essentially allowing for self-assembly
of the DNBs into DNA nanoarrays.
The next step involves Complete
Genomics' combinatorial probe-anchor
ligation (cPAL), which is an unchained,
non-iterative, base-reading sequencing
assay. According to Reid, this method has
the advantage of dramatically reducing
the required probe and enzyme concentrations
and reducing imaging time, which
substantially cuts reagent and imaging
costs, respectively. The collected images are
then assembled with Complete Genomics'
computing software.
Currently, Complete Genomics offers
sequencing at $20,000 per genome for
projects with a minimum of eight genomes,
andvolumediscountsforprojectswithmore
than24genomes.Forthoseresearcherswith
projects of even larger volume, the cost per
genome is around $5,000.
But McDougall cautions against
"comparing apples to pears" when it comes
to advancing toward the $1000 genome.
"When we [at Oxford Nanopore] speak
about the cost of the genome, we like to
consider the full cost. That means reagents,
amortization of our instruments, labor,
IT, informatics, project management,
and sample preparation," she says. Oxford
Nanopore's technology does not require
reagents, which McDougall believes gives
them an advantage when it comes to
reducing total costs.
With fame and funds abounding,
the third-generation sequencing race has
attractednumerouscompetitors.TheArchon
X-Prize for Genomics was established in
October 2006 and offers a $10-million
prize to the first team able to sequence 100
human genomes in 10 days. Research teams
and companies from around the world have
stepped up to the challenge. Participants
includeteamcrackerfromTaiwan,base4innovation
from the UK, and the US-based
companies Visigen Biotechnologies, Reveo,
ZS Genetics, and 454 Life Sciences. Also
from the US is the Foundation for Applied
Molecular Evolution and George Church's
Personal Genome X.
At the finish line
Which company will be the one to break
the $1000 genome barrier first? What will
be included in the total cost of that $1000?
How will a winner actually be decided?
Thoughthesequestionsremainunanswered,
Pacific Biosciences, Oxford Nanopore, and
Complete Genomics all predict there will
be a winner soon--and according to each
company, it's going to be them.
From the development of the Sanger
method to the completion of the Human
Genome Project, geneticists have made
significant strides in understanding and
accessing the information stored in our
genes. Whether using optics, nanopores,
nanoballs, or some yet-unannounced
sequencing methodology, the completion
of the third-generation sequencing race
will mark another milestone in the history
of genetics. Yet like all previously lauded
achievements, it will be improved upon.
Speed, accuracy, easily assembled long
reads, and reduced cost will only satisfy for
so long. Once the $1000 goal is reached,
developers are likely to set their sights on
the $100 genome, and, perhaps, someday
even the $1 genome.
References
1. Levy, S., G. Sutton, P.C. Ng, L. Feuk,and A.L.
Halpern. 2007. The diploid genome sequence
of an individual human. PLoS Biol. 5:e254.
2.Wade, C.M., E. Giulotto, S. Sigurdsson,
M. Zoli, S. Gnerre, F. Imsland, T.L. Lear,
D.L. Adelson, et al. 2009. Genome sequence,
comparative analysis, and population genetics
of the domestic horse. Science 326:865-867.
3. Margulies, M., M. Egholm, W.E. Altman,
S. Attiya, J.S. Bader, and L.A. Bemben.
2005. Genome sequencing in microfabricated
high-density picolitre reactors. Nature
437:376-380.
4.Harris, T.D., P.R. Buzby, H. Babcock, E.
Beer, J. Bowers, I. Braslavsky, M. Causey,
J. Colonell, et al. 2008. Single molecule
DNA sequencing of a viral genome. Science
320:106-109.
5.Korlach, J., P. Marks, R. Cicero, J. Gray, D.
Murphy, D. Roitman, T. Pham, G. Otto,
et al. 2008. Selective aluminum passivation
for targeted immobilization of single DNA
polymerase molecules in zero-mode waveguide
nanostructures. Proc. Natl. Acad. Sci. USA
105:1176-1181.
6.Eid, J., A. Fehr, J. Gray, K. Luong, J. Lyle,
G. Otto, P. Peluso, D. Rank, and S. Turner.
2008. Real-time DNA sequencing from single
polymerase molecules. Science. 323:133-8.
7. Clarke, J., H.-C. Wu, L. Jayasinghe, A. Patel,
S. Reid, and H. Bayley. 2009. Continuous base
identificationforsingle-moleculenanoporeDNA
sequencing. Nat. Nanotechnol. 4:265-270.
8.Rosenbaum A.M., J.V. Thakuria, X. Wu,
A.W. Zaranek, J. Li, P. Hulick, M. Murray,
M.F. Browning, et al. 2009. Clinical analysis
of individual genomes. Nature. (In press).
9.Drmanac, R. and C.A. Reid. 2009. Human
genome sequencing using unchained base reads
on self-assembling DNA nanoarrays. Science.
327:78-81.
Written by Erin Podolak.
Supplementary material for this article is available at
www.BioTechniques.com/article/113371.
BioTechniques 48:105-111 (February 2010)
doi 10.2144/000113371
Helicos Biosciences' approach to single molecule sequencing
involves binding of targets to a flow cell followed by sequential
rounds of nucleotide addition, imaging, and cleavage. Courtesy
of Helicos Biosciences.