AS TEAMS of researchers race to complete a first draft of the human genome,
two Californian companies have used some serious computer muscle to achieve
another first. They have swept through the entire sequence published so far to
turn the jumble of bases which make up our DNA into a meaningful inventory of
genes, their locations and functions.
鈥淚t鈥檚 the first step towards turning raw genetic data into methods for
diagnosing and treating disease,鈥� says John Couch, chairman of DoubleTwist of
Oakland, which has teamed up with Sun Microsystems of Palo Alto to do the
work.
Their inventory, unveiled this week, is likely to ruffle the feathers of
Celera Genomics in Rockville, Maryland, a company set up by the gene
entrepreneur Craig Venter to capitalise on analysis of information from the
human genome. Venter has been racing to sequence the genome ahead of the
publicly funded international Human Genome Project (HGP). The rivalry has been
portrayed as a battle between those who want to exploit the genome for profit
and those wishing to make it freely available.
Advertisement
Yet contrary to their uneasiness about Venter鈥檚 efforts, researchers involved
in the public programme have welcomed the new gene inventory. They see it as
鈥渁dding value鈥� to the raw data they deposit each day on the Web, where it is
available to all free of charge. 鈥淚 couldn鈥檛 be more thrilled,鈥� says Eric
Lander, director of the MIT Center for Genome Research in Boston and a senior
figure in the public effort. 鈥淭his is a real validation of the public sequencing
project,鈥� he says.
DoubleTwist used computing power provided by Sun to compile the inventory.
鈥淒oubleTwist is the first large-scale computational analysis of the human
genome,鈥� says Couch. 杏吧原创s at the company analysed all the raw sequence
data deposited so far on the Internet by the HGP. With the help of purpose-built
algorithms, they trawled through the 2 billion DNA letters looking for the
characteristic signatures of genes, such as 鈥渟tart鈥� and 鈥渟top鈥� sequences and a
lack of repeating bases, and arranged them into genes and families of genes on
each chromosome.
Their number crunching has already confirmed what many geneticists
suspected鈥攖hat we have around 100 000 genes. 鈥淲e鈥檝e found 65 000 genes
marked with high confidence, and we鈥檙e looking closely at another 40 000
potential genes,鈥� says Nick Tsinoremas, director of research at DoubleTwist.
鈥淲ithout the entire sequence, we can鈥檛 know for sure,鈥� he adds. HGP researchers
have positively identified more than 40 000 genes.
Access to DoubleTwist鈥檚 inventory comes at a price,
however鈥攁pproximately $10 000 per year for each researcher who uses
it. DoubleTwist has promised to allow cheaper access for academics, however. 鈥淚
think they will make it available on terms that make it attractive,鈥� says
Lander.
Meanwhile, officials at the Human Genome Project said this week that the
genome sequence is close to completion. The consortium announced that the
sequencing effort is now entering the home straight, and should be complete next
month. So far, researchers have assembled around 85 per cent of a 鈥渨orking
draft鈥� of the genome. 鈥淲e鈥檙e very excited about the final frame,鈥� says Lander.