Hereditary highway map: assessing the toll.

Hereditary highway map: Assessing the toll

Momentum is building among U.S. scientists to create a detailed road map of the entire human gene system, or genome. Last week, geneticists, molecular biologists and computer scientists convened in Washington, D.C., at the request of the Office of Technology Assessment (OTA) to help estimate the cost of such an undertaking--a biological mission so complex it has been likened to the 1960s effort to put a man on the moon. Congress is to consider funding for the project in the fall.

Scientists expect that human-gene mapping will lead to improved diagnosis of hereditary diseases, the development of new drugs and a host of unforeseen benefits. Enthusiasm for the project has grown in the past year with the mapping of genes responsible for muscular dystrophy and neurofibromatosis, and with the discovery that certain genetic sequences are related to manic depression (SN: 10/25/86, p.261; 6/6/87, p.359; 3/28/87, p.199). But a high-resolution map showing every human gene has only recently become feasible with the development of specialized automated technologies.

Recent advances in automation have made DNA sequencing both cheaper and faster. Until recently, according to scientists at the meeting, the cost has been $1 to $2 per nucleotide base; these bases spell out the genetic code. New technologies have lowered the costs to as little as 6^ or 8^ per base, says Leroy Hood of the California Institute of Technology in Pasadena. And within six months, he predicts, the cost could drop to a penny a base. Such differences are significant, he says, as there are approximately 3 billion bases in the human genome, and each base will have to be mapped at least two or three times to confirm its location.

Researchers at the meeting also noted progress in the number of genetic markers --key chromosomal reference points --that have been identified (SN: 8/31/85, p.140). To date, 300 to 400 "reasonably informative' markers have been identified, says Helen Donis-Keller, a senior research director at Collaborative Research Inc., a Bedford, Mass.-based biotech research laboratory. An additional 300 to 400 such markers will be needed to develop a genetic map that would have a marker every 5 million bases--a scale that would be very useful for locating the sites of disease-causing genes, Donis-Keller says. She predicts that such a map will be completed in the next two years. Detailed nucleotide sequencing, with its ability to determine exactly which proteins are coded for by defective genes, would take many more years.

Scientists say it will be necessary to develop highly sophisticated computer programs to make sense of the huge amount of genetic data that will be generated by the mapping project. It is not unreasonable to assume that a supercomputer may be needed, according to some scientists at the meeting. And millions of dollars may be needed to train specialists with combined skills in molecular biology and computer science.

How much would the gene mapping project cost? It will be some time before OTA analysts and up the numbers. But several scientists express surprise that while much of the mapping itself could be done for $100 million, the costs of simply freezing "signpost' cell lines for future use might amount to a quarter of a billion dollars or more. "At that price,' says Harvard researcher and Nobel laureate Walter Gilbert, "it would be cheaper to make the stuff all over again instead of storing it.'

If in fact every important cell line were to be cloned and stored, "it would take 12 of the largest liquid-nitrogen refrigerators now available,' says Robert E. Stevenson, of the American Type Culture Collection, a cell-storage bank in Rockville, Md. "We're talking about a large [liquid-nitrogen] tank farm.'

Total cost of the project will also hinge on the total number of human genomes mapped, says Paul Berg, Stanford biochemist and Nobel laureate: "Whose DNA are we going to sequence? Are you satisfied with one? Is that the human genome?'