ROCKVILLE, Md. -- Biologist J. Craig Venter once raced the U.S. government to complete the decoding of the human genome. Now, after a maverick career studying the code of life, Dr. Venter has a new goal: life itself.

Along with two veteran collaborators, Dr. Venter hopes to become the first to whip up a made-to-order bacterium. Normally, new life is created via reproduction, with each generation passing its genes on to the next. But Dr. Venter aims to bypass that process by manufacturing a complete set of genes, or genome, of a single-cell bacterium in his laboratory. This man-made genome would be installed inside a bacterium whose own genes have been removed.

By creating such a life form, Dr. Venter's researchers think they may come closer to understanding what life is and how scientists can manipulate it for the benefit of humankind. New artificial species could open avenues for industrial production of drugs, chemicals or clean energy.

"This is the step we have all been talking about. We're moving from reading the genetic code to writing it," Dr. Venter says, swiveling in his chair at his sprawling scientific headquarters here.

Fueling Dr. Venter's latest ambition is $12 million in grant money from the U.S. Department of Energy. An additional $30 million being raised by Dr. Venter and his business partners from several wealthy private investors will bankroll a new company, Synthetic Genomics Inc. The company will join in funding the research program at Dr. Venter's nonprofit research institute and own the rights to any intellectual property it creates.

A growing number of researchers are speeding toward similar goals. Teams at Harvard University and in Japan are attempting to make new versions of common E. coli bacteria with completely synthetic genes. On many campuses "synthetic biology" is the latest buzzword. Students are now taught to program the DNA of bacteria as if it were software running a computer.

Scientists have known for three decades how to add genes to a bacterium. That is the basic discovery behind the biotechnology industry. For example, insulin for diabetics is now manufactured by splicing an insulin-making gene into a microbe. Dr. Venter's plan is to take this technology to its logical extreme -- to manufacture and stitch together all the genes necessary for a bacterium to survive. For now, he will still need the shell of a living microbe, with its genome removed, to complete the creation of what he calls the first "human-made species."

The most basic hurdle Dr. Venter faces is getting his concoction to work. It's possible that his team could create a genome, place it in a bacterial cell, and find that nothing happens -- analogous to installing a new operating system in a computer and then being unable to turn on the computer.

If the cell does "boot up," Dr. Venter believes the creation will have "10,000 applications," providing a basic template onto which scientists could add and test new functions. He imagines an organism perfected to make clean hydrogen energy from sunlight. Another would chew up cellulose, the raw material of plants, and spit out ethanol that could be used as car fuel.

Industry is already chasing similar applications using existing techniques. Scientists at DuPont recently rewired E. coli germs with 23 separate genetic modifications so that the bugs will churn out an ingredient used in a nylon-like fiber. The company is now building a $100 million biological manufacturing plant in Loudon, Tenn. "We could talk about these ideas in 1980. Now you can actually do some of them," says John Pierce, DuPont's head of biotechnology research.

On the scientific side, Dr. Venter believes his experiment will lead to a greater understanding of how living things work. Some synthetic biologists compare themselves to engineers. What scientists have done to date is akin to taking apart existing cars and seeing what happens when a Ferrari engine is installed in a Toyota. That is useful, but an even better way to understand cars is to build one from scratch.

Predictions of super-charged microbes are cause for worry in some quarters. Several government panels are now studying whether terrorists or biological "hackers" could misuse the technology to concoct new germs or reanimate old ones, such as smallpox. "This is 'Jurassic Park' science, so it's bound to get attention, but it does not appear to facilitate bioterrorism at this time," says John H. Marburger, the White House science adviser.

Dr. Venter is best known for his role as president of Celera Genomics, a unit of Applera Corp. founded in 1998 to generate a private copy of the human genome. Taking advantage of new, speedier devices for reading DNA, the company raised nearly $1 billion from stock offerings and was able to complete the job in just two years, overtaking the publicly funded International Human Genome Project.

Similarly, new technology is now also making it easier to "write" DNA, or create it from scratch using chemical building blocks. These building blocks are sometimes called "letters" and abbreviated as A, C, G and T. Today, customized strings of DNA can be ordered from several companies. A single gene can still cost $4,000 or more. But costs are dropping rapidly. "If you look at the curve, it's headed to about zero in 2006," says John Mulligan, CEO of Blue Heron Biotechnology, a Bothell, Wash., company that sells DNA.

That's a trend Dr. Venter hopes to capitalize on. Along with a colleague recruited from Celera, Nobel laureate Hamilton O. Smith, the group is hoping to build a complete copy of the DNA of a bacterium called Mycoplasma genitalium. A denizen of the human reproductive tract, the germ is notable because it has a single chromosome with only 517 genes -- about as few as any living organism known. By contrast, humans have an estimated 20,000 to 25,000 genes.

Dr. Smith predicts it will take two years to make a synthetic copy of M. genitalium. The complete genome contains about 580,000 DNA letters, far bigger than any piece of DNA synthesized to date.

The project to create an artificial cell dates back to 1995, when Dr. Venter led the first team to sequence all the DNA of a living organism, Haemophilus influenzae. The decoded germ had 1,800 genes, yet M. genitalium, the second bacterium studied by Dr. Venter's research group, proved to have fewer than one-third as many. "That started the whole thinking of, 'How small can it be? What is the minimal definition of life?' " says Dr. Venter. "If we can't answer that question, we can't pretend to understand what life is."

For practical applications as well, a minimalist genome would be a helpful starting point because the bacteria wouldn't have any unneeded parts, saving their energy for whatever process they're being used for.

Dr. Venter and a collaborator, Clyde Hutchison, began systematically disabling M. genitalium's DNA a piece at a time, looking for colonies that could still grow despite the missing parts. From 517 genes, they pared the genome down to a must-have list of between 265 and 350.

Drs. Venter and Smith concluded the only way to test the viability of a minimal genetic life form would be to build one. But those plans got put on hold when Dr. Venter took the top job at Celera. There he battled public-sector genome researchers, many of whom weren't happy to be challenged by a for-profit company. Eventually the two sides agreed to unveil their results at the same time, in a face-saving "tie" that President Clinton announced at the White House in 2000.

The race to decode the human genome highlighted several traits of Dr. Venter: fierce competitiveness, a flair for grand undertakings and an ability to attract investors' funds.

Celera struggled to turn its scientific feat into a profitable business. Following disagreements with the parent company, Dr. Venter was ousted as chief executive in January 2002. But he wasn't out of the spotlight for long. A few months later it emerged that the human genes decoded by Celera were largely his own. The company had also decoded the DNA of his poodle, Shadow.

Putting about $150 million in stock gains from Celera and other ventures into a foundation, he built a new $35 million nonprofit J. Craig Venter Institute here, and began a voyage around the globe on his sailboat, Sorcerer II, collecting DNA samples from the world's oceans. He has also started an autobiography in which he plans to reflect on his own genetic makeup, and is now looking to hire an archivist so the institute can build a permanent record of his exploits. "A lot of people were hoping that I would retire," says Dr. Venter, who turns 60 next year.

The synthetic genome project also remained a "major personal goal," he says, and in June 2002 the institute filed a grant application to the Department of Energy to fund a trial project.

That summer, as Dr. Venter was building his research team, polio researcher Eckard Wimmer of the State University of New York at Stony Brook produced a surprise scientific advance. Using DNA strands ordered from the Internet, he and several co-workers generated new supplies of poliovirus from scratch. The project, carried out with funding by the U.S. Department of Defense's research arm, had been conceived as a way to demonstrate the security risks of synthetic DNA.

Dr. Venter was quick to insert himself in the ensuing debate. In newspapers and on television, he charged Dr. Wimmer with sloppy and "irresponsible" work. In Dr. Venter's view, creating a human pathogen was a sure way to give biology a bad name and sow public mistrust.

But Dr. Venter's team had clearly been scooped. In its grant application to the Department of Energy, it had proposed a similar experiment using a virus called PhiX 174 that is harmless to humans. Despite a second-place finish, when Dr. Venter's team eventually succeeded in synthesizing PhiX a year later, the DOE presented it in public as a major breakthrough. In a November 2003 news conference announcing the result, Energy Secretary Spencer Abraham called it "nothing short of amazing." A DOE scientific evaluation, by contrast, termed Dr. Venter's result "nothing new." Dr. Venter defends the importance of his results, saying his synthesis method was faster than the earlier one.

The work with PhiX and polio amounted to a test run, since most scientists don't think viruses are truly living things. With a hundred times more DNA than a virus, building a synthetic bacterial chromosome remains a much more difficult goal.

Earlier this year, Dr. Venter along with Dr. Smith and two business associates began planning a new company to accelerate their artificial bacteria project, believing they needed fresh funds to complete it in a reasonable amount of time. "I may not be alive to see it," says Dr. Smith, 73, half-jokingly.

The group had little difficulty finding investors, says company president Juan Enriquez, an author and executive based in Boston. About $15 million was contributed by Mexican industrialist Alfonso Romo Garza, who formerly had interests in vegetable seeds, insurance and cigarettes. A similar sum was raised from fewer than six other wealthy individuals, Mr. Enriquez says. Mr. Romo declined to discuss the size of his investment.

Dr. Venter ruled out seeking money from professional venture-capital funds, fearing they would diminish his control of the research and push too quickly for profits. After his experience with Celera, Dr. Venter said he was looking for investors "who can actually see a horizon beyond three months."

Currently, there are 30 people on the project to create a synthetic M. genitalium and related efforts, a figure likely to jump to 100 over the next 12 months. Some institute documents refer to the still-hypothetical new species as Mycoplasma laboratorium, but other names are under consideration. "Mycoplasma venterium," suggests Dr. Hutchison.