Science: Superfast DNA sequencing
The Human Genome Project took 13 years and $2.7 billion to come up with the complete sequence of human DNA. This week in the journals Nature and Science, two research groups have published the first wave of new, superfast sequencing methods; in a few years, scientists hope that these methods and others like them will let one human's DNA be sequenced for less than $100,000, ushering in a new era of data-based personalized medicine.
One new technique, published July 31 in Nature, is already commercially available as a $500,000 machine that uses a credit-card-size chip with about 1.6 million tiny wells; each well holds a tiny bead attached to 10 million copies of one short fragment of DNA. The machine sequences by washing the four bases that make up DNA—adenine, cytosine, guanine, and thymine—across the chip one at a time; if a base finds its partner in a fragment of DNA, it binds there and lets out a flash of light, which the chip's fiber optics detect.
Recently, this technique has managed to read more than 100 million bases in five hours. The method used now to sequence DNA can manage only about 67,000 bases in a one-hour run. Like computer chips, the method should get faster and cheaper all the time, says Jonathan Rothberg, founder and chairman of 454 Life Sciences Corp., the company that sells the machines. With one of his sequencers, he says, one person can do the work of a hundred people and millions of dollars' worth of robots, sequencing an entire genome in a few hours.
The other technique published this week uses equipment that many labs already have, says Harvard genetics professor George Church, whose team published the paper in Science [www.sciencexpress.org]. In their method, all four bases are marked with different colors, and are flowed across a chip simultaneously. A camera mounted on a microscope takes pictures of the different-colored bases as they attach. "Our goal is to say anybody could do this," Church says. His paper includes step-by-step instructions, although steps like "add 800ul NX2 buffer (100mM NaCl, 10mM Tris-HCl pH 7.5, 1 mM EDTA, 0.1% Triton X-100)" might be hard to follow at home—if not in the lab.
Both of these methods are part of a push by the National Human Genome Research Institute to dramatically reduce the cost of sequencing one person's genome—and improve their medical care. It's analogous to checking a patient's blood type before giving them a transfusion, says George Weinstock, co-director of the Human Genome Sequencing Center at Baylor College of Medicine. If scientists know that, say, prostate cancer patients with a particular mutation in a particular gene get better when they're given a particular cancer drug, having the sequence of that gene would help.
"Now you take all of your genes and multiply across all of the possible things we know about‑diseases, responses to medicine, issues for surgery, blood types," Weinstock says. One day, he predicts, a doctor will be able to tell what's wrong with you by plugging your genome's sequence into a computer.
advertisement
