Gene Projects Struggle To Forge Cooperation

With the race to decode the human genome nearing completion and the stocks of genome companies soaring, a last-minute merger negotiation between public and private competitors has foundered in a clash of principles and egos. In an exchange of letters over the last week and a half, each side sought to blame the other for the breakdown and for failing to negotiate in good faith.

Although a long history of disagreement between the principals on each side probably did not help matters, a basic obstacle was an impasse between the public consortium's desire to make the information in the genome freely available to all researchers and the need of the private company, Celera, for enough proprietary safeguards to make a profit.

The decoding of the human genome is expected to mark a new era in medicine in which diseases will be analyzed and treated in terms of the genes that cause or influence them.

Identifying the sequence of the 3 billion chemical units of DNA that make up the human genome has been a 10-year project of a public consortium of university centers, financed largely by the National Institutes of Health and the Wellcome Trust of London.

But in May 1998, the newly formed Celera Corp. of Rockville, Maryland, leapt into the fray with the surprise announcement that it would sequence the genome from scratch with a novel approach and finish several years ahead of the public consortium's target date.

When Celera's approach seemed to be succeeding, despite predictions from some scientists that it could never work, observers in both camps realized that they were pursuing the same end with complementary means. Celera had a top-down method for sequencing the genome, the public consortium a bottom-up approach. If the two data sets were combined, the genome might be sequenced much sooner.

Most of the DNA sequencing on the public consortium's side has been carried out under Dr. John Sulston at the Sanger Center in Britain and Dr. Robert Waterston at Washington University in St. Louis. Both strongly support the idea of placing all human DNA immediately in the public domain, and have reservations about dealing with Celera.

It was another member of the public consortium, Dr. Eric Lander of the Massachusetts Institute of Technology, who approached Dr. J. Craig Venter, the president of Celera, with the idea of a collaboration.

After preliminary talks, a meeting was held between the principals on Dec. 29. On the public consortium's side were Waterston, Dr. Francis Collins, head of the National Human Genome Research Institute, and Dr. Harold Varmus, then director of the National Institutes of Health. Venter was joined by Tony White, president of Celera's parent company, PE Corp.

But Lander and other doves were absent, resulting in a meeting of hawks without a mediator. The institutes' side insisted that the merged data set should be freely used by everyone, including Celera's commercial competitors. White indicated he could not agree to that, and the meeting broke up in mutual feelings of distrust. After two months of silence, Collins and Varmus wrote on Feb. 28 to Celera that they would assume that the idea of cooperation was dead unless they heard from Celera by March 6. The two officials included a proposed statement of shared principles, including that "the current antagonism and excessive competition should be replaced with a more collaborative spirit.''

On Tuesday, a day after the institutes' deadline, Venter replied that Celera was still interested in collaboration and that the institutes' letter had "dramatically misstated'' the company's position on its needs for intellectual property protection.

The stark disagreement is in one sense surprising because the two sides recently joined forces in successfully sequencing the genome of the laboratory fruit fly, used by Celera as a critical pilot project for its human genome strategy.

But fruit fly genes are of no direct medical value, and the two sides were able to share credit and make their data public without acrimony.

With the human genome data, there are much higher stakes in terms of intellectual credit for a landmark achievement and the prospect of a huge commercial payoff, although it is unclear exactly where the payoff will come. The genome itself is just a mystifying string of the same four chemical letters, with no evident punctuation, annotation or chapter headings. Interpreting and making use of it will be the work of decades.