Services
– ABI Ion Torrent S5
• Services provided
• How the S5 works
• Cost considerations
• Adapters
• Barcodes
• Run times
• Data analysis
• General NGS Recommendations
Critical Note: As of 31 December 2023, ThermoFisher is discontinuing sales of the kits necessary to run our S5 using the OT2 for templating; thus, starting in January 2024, we will be able to process samples only by using expired kits. As a result, researchers should contact the Genomics Core prior to beginning a project destined for the S5.
The Ion S5 combines semiconductor sequencing technology with
natural biochemistry to directly translate chemical information into digital data. The Torrent
Suite Software performs preliminary reference genome alignment and outputs data in FASTQ and BAM (aligned or un-aligned) formats for easy downstream analysis with optional third party analysis tools. Depending on the chip and kits used, final reads from a single run range from ~20-30 million (up to 400bp) or ~60-80 million reads (up to 200bp).
Services provided
Typically, libraries are prepared by the Client; however, under certain circumstances, the Genomics Core will prepare the libraries from DNA provided by the researcher. The Genomics Core performs the Library QC assay, templating (by emulsion PCR on the OneTouch·2 or Isothermal Amplification), the S5 reaction (sequencing), and production of data output files (FastQ, un-aligned BAM, or BAM; split by barcodes, if applicable). If an appropriate genome is provided to the Core, the data will be aligned on the Torrent Server using TMAP... software optimized for Ion Torrent data by incorporating flow-space information.
How the S5 works
The sequencing technology underlying the S5 exploits a well-characterized biochemical process.
When a nucleotide is incorporated into a strand of DNA by a polymerase, a hydrogen ion (H+) is
released as a byproduct. This hydrogen ion carries a charge which the S5's ion sensor
(essentially the world's smallest solid-state pH meter) can detect. As the sequencer floods the
chip with one nucleotide after another, any nucleotide added to a DNA template will be detected
as a voltage change, and the S5 will call the base. If a nucleotide is not a match for a
particular template, no voltage change will be detected and no base will be called for that template.
A principal component of the S5 is the sequencing chip. This microprocessor chip incorporates
an extremely dense array of micro-machined wells married to a proprietary ion sensor. Each well
contains a different DNA template, allowing massively parallel sequencing. Two chip sizes are
available:
400-bp or 500-bp chemistry:
- 530 chip: ~38 million wells (~20 million Final Reads).
200bp chemistry:
- 540 chip: ~148 million wells (~60-80 million Final Reads).
Because it detects nucleotide incorporation without the use of light, the S5 uses the simplest sequencing
chemistry possible – natural nucleotides. There is no need for expensive and error-prone modified bases, enzymatic
cascades, chemiluminescence, or fluorescence. Direct detection also means the incorporation of each nucleotide is
recorded in seconds. As a result, Ion sequencing and basecalling (including alignment, if providing a reference genome) can be completed overnight, even for 400bp chemistry (using emulsion PCR on the OneTouch·2 for templating) or 500bp chemistry (using Isothermal amplification kits for templating).
A more extensive explanation of the platform can be found at the manufacturer's website,
Ion Torrent S5
↵ Back to Top
Cost Considerations
Keep the following points in mind when preparing proposals for NGS
experiments, whether you do them here or send samples elsewhere. In particular, when comparing costs of performing your experiment with an Illumina-based Core vs. an Ion-based Core, ensure that you are truly comparing 'apples-to-apples'.
Do not assume your NGS experiment will work correctly the first time; thus, factor in the possibility that a run will need to be repeated with a new library (or new set of pooled libraries).
Some projects may require extensive optimization of the NGS libraries, each round of which will require another NGS run for testing.
Recognize that experiments requiring paired-end runs on an Illumina platform can often be done with ~1/2 the reads on an Ion platform, by virtue of the longer read length possible with Ion chemistry.
The stated number of reads may not reveal the full story as many of those reads might be filtered out of the final data set for various reasons; thus, investigate how many of those reads are likely to be filtered out once the data are in your hands.
With Ion data, the number of Final Library Reads includes only reads that were not already removed by filtering for polyclonal ISPs, low-quality reads, and adapter-dimers.
Factor in whether using one platform vs. the other will result in substantial delays in your project (e.g., waiting to pool sufficient samples to leverage a 'lower cost/base' platform).
In the end, the real question is how much money and time will it cost to complete your project by either platform... not the total number of reads or Gb of data produced. In some cases, time may actually be more precious than money!
↵ Back to Top
Adapters
Ion sequencing uses two adapters, with the template being attached to the ISP (i.e., the sequencing surface) by the P1 adapter and sequencing occurring from the A1 adapter. For some projects, it may be preferable to use a truncated P1 (trP1) adapter.
For low-level multiplexing, the Genomics Core can provide (at cost) aliquots of official Ion barcode adapters.
Adapters can be purchased directly from Ion, or prepared by any reputable company that synthesizes oligos... as long as the adapters are subjected to purification by HPLC to remove non-full length oligos.
We can provide clients with the current adapter sequences, as well as any recommended oligo-modifications.
Adapters (ds-oligos) can either be ligated onto DNA templates or incorporated into amplicons by designing them into the PCR primers; however, both options have their own pros & cons that should be clearly understood prior to committing to either approach.
The 3'-end of the A1 adapter cannot be altered, as it contains the critcal 'key' sequence which Ion software uses to identify true reads.
↵ Back to Top
Barcodes
We strongly recommend using Ion's barcode sequences rather than designing your own barcodes or using primer-sequences as surogate barcodes.
Ion uses 'flow space' for basecalling; thus, barcodes for which the design did not take 'flow-space' into account are likely to be sub-optimal.
Given that Ion's barcode sequences are designed to create maximum separation based on 'flow space', Ion's barcodes are less prone to being mis-assigned than barcodes which were designed purely on 'base-space'.
Ion provides for up to 384 barcodes; we can provide you with the current list of sequences.
Barcodes can be purchased directly from Ion, or prepared by any reputable company that synthesizes oligos... as long as the barcodes are subjected to purification by HPLC (to remove non-full length oligos) and the 'GAT' spacer sequence is included (which cleanly defines the end of each barcode in 'flow-space').
↵ Back to Top
Run Times
With the S5, most samples can be fully processed (library QC, templating, and S5) within 2-3 days. The sequencing reaction run time varies with read length (up to 400-bp) and chip type (520, 530, or 540).
↵ Back to Top
Data Analysis
Generally speaking, data analysis will be the Client's responsibility. However, as needed, the
Genomics Core will provide support with regard to enlisting help from its Ion contacts.
Please note that NGS experiments generate massive amounts of data, in the form of hundreds of thousands
to millions of reads. Currently, reads from the S5 are ≤400-bp long (when using the OneTouch·2 Templating kits) or ≤500-bp ( if using the Isothermal Amplification kits for templating). Thus, data analysis can actually be the hardest portion of an NGS experiment.
↵ Back to Top
General Recommendations
Next-Generation Sequencing (NGS) experiments are expensive and more complex than traditional
Sanger-sequencing. Further, analysis of NGS data is not trivial. Thus, to fully leverage the S5 platform,
you need a thorough understanding of the instrument's abilities with regard to your research question and a sound
experimental design – including how to analyze the resulting Gb of data – before preparing your templates.
Investigate the literature and discuss your research interests with knowledgeable people.
Consult with our staff regarding your proposed experimental protocol as well as the timing of your run
well ahead of when you intend to submit your samples.
Determine what software is required for data analysis. Then, download appropriate large data sets to
practice using the software and to verify that your computing resources are sufficient.
Determine what databases (if any) will be required; for example, whole genome sequencing requires
a reference genome for sequence alignment. Download the appropriate information from the required
databases in advance and ensure that it is in the appropriate format for analysis with your
chosen software.
Make your initial experiment as simple as possible, while still getting usable data. The S5 has powerful
capabilities (such as barcoding) that can lead to lower per sample costs; however, with those abilities come
other considerations such as lower depth of coverage and the need for more software analysis.
Heed the 'Garbage in, Garbage out' mantra when making your libraries and quantifying your template.
↵ Back to Top
