[Paleopsych] NYT: 2 New Methods to Sequence DNA Promise Vastly Lower Costs

[Premise Checker]

2 New Methods to Sequence DNA Promise Vastly Lower Costs
New York Times, 5.8.9
http://www.nytimes.com/2005/08/09/science/09dna.html

    By [3]NICHOLAS WADE

    A new way of decoding DNA, potentially far cheaper and quicker than
    the existing method, has been developed by researchers at the Harvard
    Medical School.

    The Harvard team, led by Jay Shendure and George M. Church, describes
    the method in the current issue of Science. Last week, another new
    method for sequencing DNA was announced by a company in Branford,
    Conn., 454 Life Sciences.

    The two methods, if they work, will represent a remarkable scaling
    down of Big Science, essentially putting the equivalent of a $50
    million genome-sequencing center on the desk of every researcher and
    physician.

    The methods are giant strides toward the goal of sequencing the human
    genome so cheaply that it could be done routinely for medical reasons.
    The rallying cry for this goal is the $1,000 genome. "The $1,000
    genome has been my passion and obsession ever since I was a graduate
    student," Dr. Church said.

    A price tag of even $20,000 or so, which now seems attainable in the
    next few years, would bring whole genome sequencing within the same
    range as other medical procedures.

    The new sequencing methods are quite similar in approach. They load
    the DNA fragments to be sequenced onto ultrasmall beads and visualize
    the sequence of each fragment through reactions that cause the beads
    to light up.

    A principal difference lies in the cost of the equipment. The DNA
    sequencing machine now being sold by 454 Life Sciences costs $500,000.
    Jonathan M. Rothberg, chairman of the board, says a single machine
    does the job of a $50 million sequencing center.

    The Harvard machine is even cheaper. It uses "off-the-shelf
    instrumentation and reagents," the authors say, explaining how
    researchers can set up sequencing centers with mostly standard
    equipment.

    The most expensive element is a $140,000, computer-controlled digital
    microscope needed to record the color changes on a slide containing
    millions of DNA-carrying beads. For labs that already possess such a
    microscope, as many do, the equipment costs would be small. All they
    need do is follow the free recipe provided by Dr. Church.

    Instead of using bacteria to amplify fragments of DNA by reproducing
    them, the Harvard method captures each fragment in a drop of liquid,
    which contains all the ingredients for the chemical amplification
    method - the polymerase chain reaction.

    The contents of each drop are loaded onto beads that are then embedded
    in a gel, with 14 million beads being packed into an area the size of
    a dime, and fluorescent chemical probes are used to indicate what the
    DNA sequences are.

    The machine developed by 454 Life Sciences uses the same amplification
    method, which was developed by Devin Dressman and colleagues at the
    Johns Hopkins Medical Institutions in Baltimore. But the beads are
    made to signal their sequence by activating luciferase, the light
    producing enzyme in fireflies , and the flashes from each bead are
    recorded by a light-sensitive chip.

    Dr. Church and Dr. Rothberg are enthusiastic about their own methods.
    Dr. Church says his method is more accurate and the equipment is far
    cheaper. Dr. Rothberg says his machines can sequence novel genomes
    whereas the Harvard method is good only for resequencing, or looking
    for variations in a genome of known sequence.

    The first human genome to be completed, by the Human Genome Project in
    2003, probably cost about $800 million. Doing a second human genome by
    the traditional methods would now cost around $20 million. The two new
    methods promise to be much cheaper. Dr. Rothberg says a human genome
    could be resequenced now by his method for $1 million. Dr. Church
    estimates that he can do a human genome for $2 million now and for
    $20,000 in the future.