Human Genome Project history 5: 'Why fiddle around?' 28/2/01. By Georgina Ferry Proposals in 1995 for a massive acceleration in sequencing of the human genome foundered amid concerns about centralisation and accuracy.

The success of the pilot worm sequencing project, says Dr Robert Waterston, now Director of the Genome Sequencing Center at Washington University, was the key to getting human sequencing off the ground.

"For the first half of the 90s," he says, "there was an underlying scepticism that something as large as the human genome could be done with gel-based approaches. People were waiting for some revolutionary new technology that would do the human genome at the touch of a button, or at least within a year. With the worm, we achieved a high enough throughput at low enough cost to convince people that it could be done with the existing technology."

The model organism sequencing projects, says Francis Collins, who succeeded James Watson as Director of the National Human Genome Research Institute in 1993, showed not only that large-scale genome sequencing was practicable, but that the information obtained was of high value.

"All the mammalian biologists watched with envy as they saw what could be done with the yeast genome, and the worm was even more valuable," he says. "It was a big shot in the arm for those who wanted to scale up the human sequencing effort."

In 1993 the US agencies launched a new five-year plan, in which they argued that "priority should be given during the next five years to increasing sequencing capacity by increasing the number of groups oriented toward large-scale production sequencing."

But initially progress was slow, with most investment still going into mapping and technology development rather than producing human sequence. In 1995, Bob Waterston and John Sulston began to argue publicly that the biomedical community would be better served by a massive increase in investment in sequencing, coupled with a small relaxation (from 99.99 to 99.9 per cent) in the accuracy of the product, which would be selectively finished later.

Based on their experience with the worm, they said that just three centres scaled up to process 84 000 'reads' (DNA fragments a few hundred bases long) per week could cover 99 per cent of the genome to an accuracy of 99.9 per cent by 2000 - five years ahead of the target for completing the project. As Dr Sulston said at the time, if it can be done, "why fiddle around?"

But the idea had plenty of detractors, not least among those labs that had been involved in mapping chromosomes and feared that the chance to move to sequencing would be snatched from them if the few labs that were further ahead - principally the Genome Sequencing Center and the Sanger Centre - did the whole thing in bulk.

"There was also a genuine concern," says Dr Waterston, "that if we went too fast we would never end up with a truly archival product - that biology would lose the chance to put itself on a completely firm footing." These arguments won the day. Systematic human sequencing began on a pilot basis in 1995, and the proposal for a massive acceleration fell into abeyance.