Crash Project to Sequence the Human Genome

A DNA sequencing maverick is joining with the world's largest maker of automated sequencing machines to form a new company that they say will "substantially complete the sequence" of the entire human genome in 3 years. But many experts are skeptical whether the bold plan, unveiled on 9 May by Craig Venter and Perkin-Elmer Corp. of Norwalk, Connecticut, can meet the promised cost and quality goals.

If successful, this for-profit venture would duplicate the work of the $3 billion Human Genome Project, a government-financed coalition of nearly a dozen academic and contract centers--and do it faster and for a fraction of the cost, perhaps as little as $300 million. Venter aims to pull off this feat by taking a brute-force approach, tackling the entire genome at once rather than in the painstaking, piecemeal fashion currently being followed by the Human Genome Project.

The business plan calls for full-scale genome sequencing to begin next April, after Perkin Elmer has delivered 230 of a new generation of advanced machines that are currently in the testing stage. The strategy employs a technique called whole-genome shotgun sequencing, an approach that has been used to sequence several bacterial genomes. Venter expects to churn out a whopping 100 million bases of sequence data per day, every day. Within a year, he says, "we will have done 99% of the genome." The following 2 years will be spent fitting pieces together and filling key gaps.

Perkin-Elmer officials say they expect to profit from patents that the new company might obtain on rare genes for which clear biomedical uses have been identified. It will also create a whole-genome database that it will market to academic researchers and companies on a subscription basis. Moreover, because their approach will use DNA from many individuals, sequencing should reveal variations in DNA between individuals that could be valuable for use in clinical research and drug testing. The company will put together a proprietary set of about 100,000 of these single nucleotide polymorphisms, says Venter.

This bold plan has split the genome community. "It strikes me that this is a cream-skimming approach," says Robert Waterston, head of a large sequencing project at Washington University in St. Louis. "It's clearly an attempt to short-circuit the hard problems and defer them to the [research] community at a very substantial cost." Others welcome the venture as a quick way to get the most useful sequence data. "I think it's great," says David Cox of the Stanford Human Genome Center.