Next-Generation Sequencing
Next-generation sequencing has taken the biological sciences by storm. It is a modern approach to sequencing developed from the mid-2000s onward that allows rapid sequencing of entire genomes. To understand next-generation sequencing, it is better to first understand conventional sequencing techniques. Next-generation sequencing is a catch-all term for a large variety of sequencing methods. The most widely used methods are known as “sequencing-by-synthesis” because the sequencing reaction requires DNA to be made (in a mechanism similar to DNA replication) against the strand in question (strand 1), and the DNA sequence is recorded by a computer as the strand 2 is being formed.
There are many types of sequencing methods and reactions. I have explained the three most widely used and best known methods here:
- Illumina (Solexa) sequencing
- Roche 454 sequencing
- Ion torrent: Proton sequencing
Basics
In all methods DNA is first extracted from the cell in question and this DNA is broken down either by enzymatic lysis or sonication. This is because DNA is easier to sequence in short fragments (often known as shotgun sequencing). From here, the short pieces of DNA have adapters attached to the 3' and 5' end (library preparation). This allows the short fragments to be identified following sequencing. Following these initial steps, each technology has it’s own unique sequencing method. Powerful computer algorithms then take the data, and like a jigsaw, piece the DNA sequence together until you have one sequence.
1. Illumina (Solexa) sequencing
Once the library has been prepared, the user is left with a short DNA fragment with adapters on either side. These adapters contain distinct regions including a) the sequencing binding site b) indices (a unique identifier code) and c) regions that allow the DNA to attach to a flow-cell.
A flow-cell is glass surface where specific bits of DNA (oligos) are physically attached. These oligos are complementary to region © of the adapter and therefore the oligos “capture” the DNA to be sequenced and immobilise it to the glass slide. It should be noted that there are two types of immobilising regions there is type A on one side and type B on the other (this will become clear).
So let’s think of a flow-cell as a glass slide with the immobilising region that is complementary to the one added to our DNA physically bound to it. This immobilising region is basically the same as our DNA, except it is missing everything except the immobilising region (bear with me here).
We then add our DNA sample which is single stranded. When we add our DNA, the complementary regions bind together, but our DNA is still not physically bound to the cell. To fix this, a DNA polymerase is added which initiates DNA replication. This means that our single stranded DNA now becomes double stranded with one immobilising region bound to the cell and the other is not. The strands are separated and a buffer is run over the cell which gets rid of all unbound DNA, leaving only our replicated DNA sequence physically attached.
Since the DNA has two immobilising regions and one is unbound, it now reaches across and binds to another region bound to the cell. The double strand that is separated leaving two separate but identical strands of DNA bound to the cell. This cycle is repeated over and over until there are millions of copies of the same DNA bound to the flow cell.
Once this is done, and due to the two types of immobilising regions, all reverse strand DNA can be washed away leaving only DNA in one direction. Each strand is synthesised again, but this time the nucleotides (A, T, C, G) that are added are fluorescently labelled. As each one is added, the labels are excited by a laser and they release a specific colour for each base. A computer captures this colour and is able to read the DNA as it is being synthesised.
The cycle is repeated again only this time the forward strands are removed and the reverse strands are sequenced.
Finally, powerful computer programs can analyse this data to give what is known as a “consensus sequence”. Remember that initially your DNA sample was broken into tiny bits. This happens at random so not all sequences were the same when they were read by the fluorescent labelling. However, there are overlaps which when pieced together allow the computer to give back one complete sequence.
2. Roche 454 sequencing
Again DNA extracted and broken down by lysis or sonification and adaptors are attached to the ends. As with Illumina, the DNA is made into single strands by heating up to separate the strands. Adaptors are added to each end of the DNA.
Tiny beads are added to the mix and the beads contain complementary regions to the adaptors, allowing the DNA fragments to bind to the beads.
Following, emulsion oil is added which forces the beads to separate. The beads and oil are added into microreactors containing DNA polymerase, primers, buffers and dNTPs in order to allows for emulsion PCR to occur, in which the fragments of DNA bound to the beads are amplified millions of times.
The beads are separated into picotiter plates with each bead fitting into one well. It is now that the sequencing occurs, in a process known as pyrosequencing.
Firstly dNTPs are added one by one in order. If there is no incorporation by the polymerase, the dNTPs are washed away and the next group is added. Once a dNTP is incorporated, pyrophosphase is released which is converted to ATP by sulfurylase. Next, luciferin is oxidized with ATP by luciferase and this results in a light signal. The light signal is read by a computer and a concensus sequence is formed.
3. Ion Torrent
Similar to the Roche 454 method, DNA is broken down, attached to beads and emulsion PCR is utilised to create millions of copies of the DNA to be sequenced.
What differs in the Ion Torrent method is this. During dNTP addition by DNA polymerase, a hydrogen ion (H+) is released. The Ion Torrent sequencing technique uses a chip with tiny wells in which a single bead can fit. This well is flooded with a solution containing dNTPs every few seconds, and if a dNTP is added, an ion is released and the pH of the solution changes. This pH change is measured by instruments below the chip. Since each solution of dNTPs added contains only one type of dNTP at a time, the computer knows which base has been added when the pH changes.
This is a much cheaper and faster method of sequencing, but like all methods, has its advantages and disadvantages.
