Bioengineer Frances Arnold wants to make microbes into the chemical factories of the 21st century. But she learned long ago that her designing skills pale alongside the handiwork of evolution. So like any good entrepreneur, she decided to hire the competition.

“This is a very rich and intricate composition, the code of life, and we don’t know how to write music like this,” she said. “We’re just learning how to hold a pencil when it comes to comparison with the code of life.”

Those who direct the power of evolution successfully, she told her audience at the 2011 AAAS Annual Meeting, can create new medicines, new fuels, and new materials in a sustainable and efficient fashion.

In an era where a letter of the genetic code costs less than a dollar and you can get your DNA delivered by FedEx, a do-it-yourself approach to designing new molecules is tempting. But Arnold cautioned that “we don’t know how to write the DNA that will reprogram a cell so that it can solve human problems.”

“So we can synthesize essentially any sequence we want, and the cells will start following any stupid directions we give them,” she said, “but we don’t really know what sequence will convince these cells to stop doing what they do really, really well…which is grow and divide.”

“If we want to convince them to do something else like make a gasoline replacement, “ she continued, “we don’t know how to write that code and make it very efficient.”

Evolution already knows how to write those efficient, useful codes, she said, so it made sense to discover a way to harness its capabilities.

The method Arnold and many others use is called directed evolution, and it starts with the realization that the physical number of ways to string together amino acids into a protein is far larger than the number of particles in the universe.

“But I can tell you categorically that most of those sequences are total junk…they don’t encode anything at all that’s useful, much less a molecule that you need for your microbial factory product,” Arnold said.Luckily, evolution has already performed a little quality control, she noted, and “we can literally scrape off the bottom of our shoes an ocean of molecules that nature discovered.”

Other functional molecules exist just a few mutations away from these natural products, but the trick is nudging the natural molecules toward these new possibilities. Arnold and others cultivate these mutated molecules, examine their possibilities, and control which molecules go on to parent the next generation. The accumulation of mutations over time, she explained, gradually leads the molecules “to acquire the properties that you want.”

It’s a low-tech process that could be done in a high school lab, she said, but the possibilities are exciting. Already scientists have created a palette of fluorescent protein colors for marking biological tissues, industrial enzymes to break down biomass for fuel, viruses that bind to metals and magnetic materials to build the nanowires and long-lived batteries of future electronics.

Researchers are also expanding their toolbox by creating new amino acids out of four instead of the usual three genetic letters, new cellular machinery to create proteins from these new amino acids, and polymers built with novel molecules.

This new world of bioengineering draws inspiration from multiple disciplines, an approach that complements Arnold's diverse research background. She received a bachelor's degree in mechanical and aerospace engineering from Princeton University and a Ph.D. in chemical engineering from the University of California, Berkeley. Arnold is now the Dick and Barbara Dickinson Professor of Chemical Engineering and Biochemistry at the California Institute of Technology.

Next month she will accept the Charles Stark Draper Prize, one of engineering's highest awards. Still, after such success, Arnold remains fascinated by the possibilities of directed evolution.

"Therein lies the solution to the energy crisis, the cure to cancer," she said. "There's a lot of really interesting bioengineering possibilities here, and I would like to play in that sandbox."