"This year's prize is about taking the chemical experiment to cyberspace," said Staffan Normark, the academy's secretary.
All three scientists became U.S. citizens. Karplus, an 83-year-old U.S. and Austrian citizen, splits his time between the University of Strasbourg, France, and Harvard University. The academy said Levitt, 66, is a British, U.S., and Israeli citizen and a professor at Stanford University School of Medicine. Warshel, 72, is a U.S. and Israeli citizen affiliated with the University of Southern California in Los Angeles.
Levitt told The Associated Press the award recognized him for work he had done when he was 20, before he even had his PhD.
"It was just me being in the right place at the right time and maybe having a few good ideas," he said, speaking by telephone from his home in Stanford, California.
"It's sort of nice in more general terms to see that computational science, computational biology is being recognized," he added. "It's become a very large field and it's always in some ways been the poor sister, or the ugly sister, to experimental biology."
Jokingly, he said the biggest immediate impact of the Nobel Prize would be his need for dance lessons before appearing at the Nobel banquet.
"I would say the only real change in my life is I need to learn how to dance because when you go to Stockholm you have to do ballroom dancing," Levitt said. "This is the big problem I have right now."
Warshel, speaking by telephone to a news conference in Stockholm, said he was "extremely happy" to have been woken up in the middle of the night in Los Angeles to find out he would share the $1.2 million prize and looks forward to collecting it in the Swedish capital.
"In short, what we developed is a way which requires computers to look, to take the structure of the protein and then to eventually understand how exactly it does what it does," Warshel said.
When scientists wanted to simulate complex chemical processes on computers, they used to have to choose between software that was based on quantum physics, which applies on the scale of an atom, or classical Newtonian physics, which operates at larger scales. The academy said the three laureates developed computer models that "opened a gate between these two worlds."
While quantum mechanics is more accurate, it's impossible to use on large molecules because the equations are too complex to solve. By using quantum mechanics only for key parts of molecules and classical physics for the rest, the blended approach delivers the accuracy of the quantum approach with manageable computations.
Working together at Harvard in the early 1970s, Karplus and Warshel developed a computer program that brought together classical and quantum physics. Warshel later joined forces with Levitt at the Weizmann institute in Rehovot, Israel, and at the University of Cambridge in Britain, to develop a program that could be used to study enzymes.
Jeremy Berg, a professor of computational and systems biology at the University of Pittsburgh, said the winning work gives scientists a way to understand complicated reactions that involve thousands to millions of atoms.
"There are thousands of laboratories around the world using these methods, both for basic biochemistry and for things like drug design," said Berg, former director of the National Institute of General Medical Sciences in Bethesda.
Many drug companies use computer simulations to screen substances for their potential as medicines, which lets them focus their chemistry lab work on those that look promising, he said.