Yeast, when left to its own devices, eats sugar and secretes alcohol. Amyris has engineered it to eat sugar and secrete hydrocarbons instead.

That, in a nutshell, is the business plan for the company, a synthetic biology specialist that spun out of UC Berkeley. Through selective breeding and genetic engineering, the company has created a group of organisms that effectively release industry-standard fuels, chemicals or fuel precursors when they go to the bathroom.

While several companies are tinkering with microbes to get them to produce valuable chemicals or fuels, the metabolic swap Amyris has performed – hydrocarbons for alcohol – gives it major advantages, company execs said during a press conference at its Emeryville, California facility to celebrate the opening of a prototype production facility there.

Amyris, for instance, will be able to produce a form of diesel that it will sell at the wholesale level for $2 a gallon or less, or around the same price as conventional fossil diesel, said CEO John Melo.  

"It will be around the same price as regular petrol diesel, but it will produce 80 percent less greenhouse gases, provide a 10 percent reduction in NOx (nitrogen gases) and provide the same or better performance," Melo said. "And with zero sulfur."

By contrast, standard biodiesel produced from palm or other plant oils needs a $1 per gallon subsidy to remain competitive. Nonetheless, Amyris also qualifies for the $1 a gallon subsidy, making it even cheaper than either standard diesel or biodiesel. Regular biodiesel also cuts mileage by around 10 percent.

The company's jet fuel, which will replace kerosene-based fuels, will produce 90 percent fewer greenhouse gases than the regular stuff without denting performance or mileage, he said.

The big test for Amyris will arrive in about two years. The company has created joint ventures in Brazil to create biorefineries on sugar plantations where genetically engineered yeast will feast on freshly harvested sugar (see Sugarcane Biodiesel Heads to Brazil). The resulting fuel will then be loaded onto ships and brought to the U.S. By 2010, Amyris hopes to be producing 200 million gallons a year out of its first plant and erecting more plants.

Melo also pointed out that because Amyris isn't producing ethanol (an alcohol) in Brazil but a hydrocarbon (a molecule includes hydrogen and carbons), the ethanol tariff on Brazilian ethanol doesn't apply.

The company is also working with the state of Alabama to build sugarcane plantations and yeast-powered fuel refineries in that state.

The next few years will likely be exciting ones for the future of fuel. Several startups have received millions in VC funds to bring a green transportation fuels to market. Each seems to have its own magic formula and, as critics and competitors like to point out, its own drawbacks. Mascoma, out of Dartmouth, says it has microbes that can efficiently transform wood chips and grass into ethanol. Several say they can produce fuel out of algae. Others like Range Fuels are eschewing microbes entirely and making fuel out of cooking plant matter. ZeaChem and Coskata, meanwhile, combine biological and thermochemical processes.

The big drawback for Amyris could be in the gallons of fuel it gets per acre. Melo said that the company could produce 600 to 800 gallons of fuel per acre. That's far better than corn ethanol and somewhat on par with sugarcane ethanol. However, it is lower than what cellulosic ethanol and algae growers say they will get. Some algae companies claim they will get 5,000 gallons per acre per year.

Melo's reply? "We think it's best to see people who operate at scale," he said. In other words, they will have to prove it. Of course, virtually none of them, Amyris included, have proved they can mass produce their fuels.   

The company grew out of research conducted by Jay Keasling and Jack Newman at Berkeley. The two were pioneers in the field of synthetic biology, which is the art/science of decoding  natural processes in a lab and then trying to recreate them. Why can abalone create hard shells out of plain chalk? Synthetic biologists try to figure out questions like that.

The first product out of the company was an artificial version of artemisinin, an antimalaria drug. The Bill and Melinda Gates Foundation funded the project. Synthetic biology relies heavily upon the real and computerized genetic modification techniques, including inserting genes from one species into another that have been developed in the last several decades in the pharmaceutical industry.  

"Some people refer to it as metabolic engineering on steroids," Newman said. "There are a few things we have adopted from plants. Plants are great for making hydrocarbons."

Like most alternative fuel companies, Amyris is more interested in selling and making fuel than licensing its technology. In licensing deals, large, established companies pay royalties to small startups for their inventions. While it works in biotech, the legal tangle that is licensing often results in resentment in other fields.

"Licensing doesn't capture a lot of value. The person who sells a product will capture more," Melo said.

Licensing, however, does have the advantage of freeing up inventors from worrying about distribution, sales, marketing and logistics. Big companies can typically do that far more economically. So how will Amyris get around that? Rather than work with major oil companies, Amyris will sell its fuel directly to large customers – i.e., retailers with centralized trucking operations – similar to the way Dell sold PCs to corporate America. Concentrating on a smaller number of large customers cuts down many of the distribution headaches.

"It is not like we are going to be selling it to gas stations," Melo said. 

Join industry leaders and influencers at Greentech Media's new conference series Greentech Innovations: End-to-End Electricity on November 17 and 18 in New York City.