The Personal Genome Machine (PGM™) Sequencer is a benchtop PostLight™ semiconductor-based platform that performs a sequencing run in approximately two hours. With a small benchtop footprint and a weight of 65 lbs (~30kg), low cost sequencing reagents, and simple touchscreen user interface, the Personal Genome Machine (PGM™) Sequencer brings next generation sequencing to every lab. The PGM™ Sequencer requires a Torrent Server for operation. The PGM™ includes a one year warranty on parts, labor, and travel for service personnel. A field service installation plan for the Personal Genome Machine System maximizes your productivity with installation of the Personal Genome Machine™ System by one of our certified field service engineers. For more info visit http://products.invitrogen.com/ivgn/product/4462917?ICID=search-product?CID=PGM-SS-12312
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Ion semiconductor sequencing uniquely enables both accurate long reads and paired-end sequencing
1. APPLICATION NOTE
Paired-End Sequencing
Ion semiconductor sequencing uniquely enables
both accurate long reads and paired-end sequencing
PANEL A
Summary
• Ion Torrent recently launched a
long read kit for the Ion PGM™
sequencer with modal high-quality
read lengths of 225 bases.
• Read lengths greater than
500 bases are feasible, as
demonstrated by the generation PANEL B
of a perfect 525 bases read.
• A novel paired-end sequencing
(PES) protocol for the Ion PGM™
system has been demonstrated.
Ion PES further enhances
accuracy. The number of indel
errors was reduced 5-fold, and the
total number of consensus errors
was reduced 10-fold.
Figure 1. Ion semiconductor sequencing utilizes natural nucleotides and simple sequenc-
ing chemistry, producing accurate reads that are at least twice as long as reads generated
with fluorescent sequencing-by-synthesis (SBS) chemistry. Sequencing accuracy generally
decreases as read length increases, but this slope is relatively flat for the Ion PGM™ sequencer
when compared with fluorescent SBS chemistry. The Ion PGM™ sequencer demonstrates modal
read lengths of 225 bases (panel B). Measured per-base accuracy (Phred score) is calculated after
aligning reads to the DH10B reference genome and is then plotted as a function of position in read,
(panel A). A Phred score of 20 is equivalent to 99% accuracy, or 1 error per 100 bases.
Next-generation sequencing — based applications and methodologies are
extensively used to investigate genome biology. The lengths of next-generation
sequencing reads are limited by accuracy—as read length increases, accuracy
decreases. This fundamental inverse relationship between read length and
accuracy governs sequencing-based applications.
2. Single-Well Ionogram for (2406, 1991)
CGCTAAGTAATATTCGCCCCGTTCACACGATTCCTCTGTAGTTCAGTCGGTAGAACGGCGGACTGTTAATCCGTATGTCACTGGTTCGAGTCCAGTCAGA
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
CGCTAAGTAATATTCGCCCCGTTCACACGATTCCTCTGTAGTTCAGTCGGTAGAACGGCGGACTGTTAATCCGTATGTCACTGGTTCGAGTCCAGTCAGA
1 100
GGAGCCAAATTCTAAAAATTCGCTTTTTTAGCGCAATGTCACTGACCTTAGTTGAACATTGTTTTTTAACGGATAGCGGGTTTTTAACATCTTAAGCGCC
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
GGAGCCAAATTCTAAAAATTCGCTTTTTTAGCGCAATGTCACTGACCTTAGTTGAACATTGTTTTTTAACGGATAGCGGGTTTTTAACATCTTAAGCGCC
101 200
CTCGACCTTTATGGTTGAGGGCGTTTTGCTATGAACGCCATCACCATTTTCCCCTCGATTATAAAACTTGAGTTATTCAGTAGTCTCCCCTCTTGCAACT
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
CTCGACCTTTATGGTTGAGGGCGTTTTGCTATGAACGCCATCACCATTTTCCCCTCGATTATAAAACTTGAGTTATTCAGTAGTCTCCCCTCTTGCAACT
201 300
CACACCCAAAACTGCCTAACGAAAAGTTATTAATTTTCAATCATATTGCTATCAGTATTTACATTTTTTCGCTGTGCTAGAAAGGGCGCATTTATGTTAG
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
CACACCCAAAACTGCCTAACGAAAAGTTATTAATTTTCAATCATATTGCTATCAGTATTTACATTTTTTCGCTGTGCTAGAAAGGGCGCATTTATGTTAG
301 400
CTCGTTCAGGGAAGGTAAGCATGGCTACGAAGAAGAGAAGTGGAGAAGAAATAAATGACCGACAAATATTATGCGGGATGGGAATTAAACTACGCCGCTT
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
CTCGTTCAGGGAAGGTAAGCATGGCTACGAAGAAGAGAAGTGGAGAAGAAATAAATGACCGACAAATATTATGCGGGATGGGAATTAAACTACGCCGCTT
401 500
AACTGCGGGTATCTGTCTGATAACT
|||||||||||||||||||||||||
AACTGCGGGTATCTGTCTGATAACT
501 600
Figure 2. Demonstration of a 525 base perfect read on the Ion PGM™ Sequencer. The “ionogram” (left) displays the number of each base (A, T, C,
or G) called at each nucleotide flow during an Ion PGM™ sequencing run. The sequence alignment (right) displays the read generated by the
Ion PGM™ sequencer in the top strand and the reference genome sequence in the bottom strand.
In October 2011, Ion Torrent across the length of the read is 99%, Reads longer than 500 bases are
launched a long read kit for the PGM™ (Q20) and consensus accuracy is achievable, as demonstrated by the
sequencer, which can provide modal 99.999%, (Q50). This data is available generation of a perfect read spanning
high-quality read lengths of 225 for download at ioncommunity. 525 contiguous bases (Figure 2).
bases (Figure 1). Mean raw accuracy iontorrent.com — run name STO-409.
Paired-end sequencing (PES) was
initially developed for short-read
Step 1: Generate “forward”
sequence reads
Step 2: Prepare template
for “reverse” sequence reads
Step 3: Generate “reverse”
sequence reads
next-generation sequencing tech-
C
H nologies to increase both coverage of
E
M
the human genome and the fraction
I of generated reads aligning to the
C
A genome. Fortuitously, PES has
L
become useful for applications that
demand ever longer reads. Increasing
Torrent Analysis Suite
base calling
Step 4: Merge “forward”
and “reverse” FASTQ
Torrent Analysis Suite
base calling the length of accurately called bases
FASTQ FASTQ FASTQ can obviate the need for PES in many
forward Merged reverse
instances, allowing more of the
D
I
Bidirectional
“paired” reads genome to be accurately sequenced,
G
I Fully and for some especially demanding
T Overlapping
A applications there is no substitute for
L Partially
Overlapping long reads. Given the utility of both
accurate long reads and PES, Ion
Non-
Overlapping Torrent has recently developed
solutions for both methods.
Figure 3. Only Ion Torrent Personal Genome Machine™ (PGM™) semiconductor sequencing
technology delivers the unique combination of long reads and PES. This figure depicts the
chemical and digital workflow of the novel PES method developed by Ion Torrent. Forward and
reverse reads are denoted by green and orange colors, respectively. The degree of paired-read
overlap is shown aligned to a reference genome.
3. Using the Ion sequencing kit Each microwell on the Ion Chip, and PES methodology can produce fully
launched in July 2011, which sup- thus, each template is in perfect reg- overlapping forward and reverse
ports read lengths of 100 bases, a ister with each transistor, ensuring reads with either a 2 x 100 or 2 x 200
novel PES method was developed for that the linkage between forward and Ion semiconductor sequencing run.
the Ion PGM™ sequencer that pro- reverse reads is unbreakable even Fully overlapping reads are useful
duces paired reads, each 100 bases when the Ion Chip is removed from for further enhancing accuracy. In
in length (2 x 100). We anticipate that the Ion PGM™ sequencer between counting-based applications such
applying recent improvements in the forward and reverse sequencing as RNA-Seq or ChIP-Seq, counting
read length to the Ion PES protocol runs. Furthermore, because the Ion accuracy is improved by partially
should produce paired reads of 200 Chip directly translates biochemi- overlapping reads (Figure 3), which
bases in length (2 x 200), with the cal signals into digital signals, via increase the number of uniquely
possibility of paired reads up to 400 perfectly arrayed transistors, only mapped reads and identify redundant
bases in length (2 x 400). a particle-immobilized template, reads by demarcating the start
primer and polymerase are required and end of the sequence reads.
Until recently, no sequencing technol- in each Ion Chip microwell. This Non-overlapping reads (Figure 3)
ogy could simultaneously offer the simplicity facilitates the occurrence of are useful for detecting genomic
benefits conferred by both long reads nucleic acid biochemistry and enzy- rearrangements or de novo assembly
and PES. The Ion Torrent Personal mology directly in each microwell, because the distance between the
Genome Machine™ (PGM) semiconduc- enabling PES as well as other novel forward and reverse reads can be
tor sequencing technology is now the sequencing methods. longer than the length of a single
only platform that provides this unique read and the distance is known and
combination of long reads and PES. uniformly distributed.
Ion PES data analysis
To compare the accuracy of fully
Ion PES workflow Forward and reverse sequencing overlapping paired reads to individual
produces two FASTQ-formatted files. reads, the proper pairs were either
The simplicity of semiconductor Merging the two FASTQ files produces aligned as individual reads or as
sequencing allows a template to both unidirectional (singleton) and a single consolidated read to the
be sequenced in both directions bidirectional reads. Unidirectional DH10B reference genome using
using a single Ion Chip. The first reads are defined as Ion Chip the TMAP alignment software in
read results from the polymerase microwell positions that do not the Torrent Analysis Suite. A total
extending the primer towards the intersect in the forward and reverse of 977,458 reads were paired and
Ion Sphere™ particle, in the “forward” FASTQ files, whereas bidirectional
direction (Figure 3, step 1), the reads are defined as the intersection
standard operating procedure for of the Ion Chip microwell position TGACGATTGC
the Ion PGM™ Sequencer. Then, off in the forward and reverse FASTQ TGACG-TTGC
instrument, directly within the Ion files. After alignment to a reference TGACGATTGC
Deletion
Chip, the template is prepared for genome, the bidirectional or “paired”
the second read (Figure 3, step 2) reads can be further subclassified CAGCT-CAGA
via a series of simple enzymatic into “improper pairs”, those with CAGCTGCAGA
CAGCTCAGA
steps. The forward primer is fully improper orientation or improper
Insertion
extended to the Ion Sphere™ particle, distance between forward and reverse
and then the original template is reads, and “proper pairs”, those CCAGGTACAC
nicked and degraded to produce with proper orientation and distance CCAGGCACAC
CCAGGTACAC
a primer for the second read. The between forward and reverse reads
Substitution
same Ion Chip is reintroduced to the (Figure 3).
Ion PGM™ Sequencer, and the second Figure 4. Depiction of a paired-read error-
read is generated by the polymerase PES generates differing degrees of correction schema. Green and orange bars
extending the primer away from the overlap between the forward and represent forward and reverse reads derived
from fully overlapping paired reads, aligned
Ion Sphere™ particle, in the “reverse” reverse reads, depending on the size
to a reference genome. Information from
direction (Figure 3, step 3). of the library fragments and read paired strands at the same nucleotide as well
lengths. With a library insert size as per-position quality scores can be used to
of 100–200 bases in length, the Ion develop improved bioinformatics methods for
error correction