THE year is 2010. You walk into a lab in a shopping mall, pay $1000 and give a swab of skin cells from your cheek. A few days later you receive an email. It is a string of 3 billion DNA letters that hold the key to your future health.
That personal genome sequencing is within sight is remarkable. The first human genome sequence took $800 million and 11 years to complete. The 鈥淪anger鈥� method used to do it has changed little since it was developed in the 1970s by Fred Sanger at the University of Cambridge. Costs have come down, but sequencing large genomes is still the preserve of large, well-funded labs.
Now two groups have developed techniques that are already 100 times faster and are set to get faster and cheaper still. The new methods do away with the need for the bacteria used in the Sanger method. Instead they use picolitre-volume beads that contain a mix of the chemicals needed to amplify DNA letters or bases. These beads are so small that millions of them can be built into a single 鈥渃hip鈥�, allowing huge amounts of DNA to be analysed at once.
Advertisement
Jonathan Rothberg, founder of 454 Life Sciences in Branford, Connecticut, and his team demonstrated their method by sequencing the 580,000-base genome of the bacterium Mycoplasma genitalum to an accuracy of 99.96 per cent in just four hours (Nature, DOI: 10.1038/nature03959).
Rothberg鈥檚 70-by-75-millimetre chip can sequence over 200,000 fragments simultaneously. The 454 method uses a version of a technique called pyrosequencing (see Diagram). First the genome to be sequenced is cut into fragments of around 100 bases. Each bead contains one fragment which is copied many times.
Primer sequences bind to the fragments and one of the four nucleotides (A, C, T or G) is added to the mix. If the nucleotide complements the base at the first position of the unknown fragment, it releases a pyrophosphate molecule that stimulates the firefly enzyme luciferase to generate a flash of light. The nucleotides are washed away and the cycle is repeated with the next nucleotide.
鈥淐ost will fall fast as we shrink our chips,鈥� says Rothberg. He is predicting that miniaturisation will follow a version of 鈥淢oore鈥檚 law鈥� 鈥� the rule of thumb that says computer chips halve in size every 18 months.
At Harvard University, Gregory Porreca and his colleagues use a similar method, but at every stage four lengths of DNA are added, each nine bases long and with one of the four nucleotides at a known position. The nucleotides are labelled with fluorescent markers of different colours. If the nucleotide binds to the fragment DNA, it gives off a flash of light.
The Harvard method is cheaper because it uses off-the-shelf equipment, but is so far only able to 鈥渞esequence鈥� known genomes. So the method can be used to spot genetic variations that predispose a patient to cancer, for example, but it cannot be used to sequence unexplored genomes (Science, DOI: 10.1126/science.1117389).
Cheap sequencing will give individuals and their doctors the option of consulting their DNA. 鈥淭he patient will then know what diseases he or she is genetically predisposed to and will be able to make proactive life choices to avert them,鈥� Porreca says. 鈥淭he patient will also be able to receive treatment tailored to his or her own genetic make-up.鈥�
Both techniques have a great deal of potential, says Daniel Turner at the Wellcome Trust Sanger Institute in Cambridge, UK, one of the centres that sequenced the human genome. 鈥淭he main limitation seems to be that the error rates of both methods are currently relatively high for individual reads.鈥�
