Software Every Bioenergy Firm Needs

A synthetic biologist engineers program to test DNA sequences in superbugs
| August 22, 2011

Howard Salis has proof that he was paying attention during his undergraduate and graduate days—and the biofuels industry should be thankful. Salis, now a synthetic biologist at Pennsylvania State University, has developed a DNA compiler program that can design and predict DNA sequences in microorganisms. Part of his chemical engineering background and time spent in the classroom, Salis says, gave him “this idea that we solve equations and we solve them on the computer during our undergraduate and graduate curriculum.” As Salis says, those classes came in handy, and for him it meant that he was able to construct a computer software program from scratch that can design new DNA sequences to improve the efficiency of the superbugs used in the biofuels industry today.

“The software makes specific predictions about which specific DNA sequences will give you a certain amount of protein, and then we measure that,” he says. “We compare the prediction with the experimental data.” After testing a DNA sequence on the computer, the team then tests those predictions at Salis’ lab. A user can simply tell the program the amount of protein an organism should make, and the program will in return tell the user the DNA sequence that will produce the desired amount of protein. This function, Salis says, is crucial to the biofuels industry. “There have been multiple companies that used the software,” he says. “In one example, a company had spent three years working on a product. They had a bottleneck. They used the software to eliminate that bottleneck and it immediately increased production by 50 percent.”

He also points to other companies such as LS9 and Amyris that he says could benefit from using the software. “If they want to make more biodiesel or octane, they need to go through a process of mutating their DNA and then measure how much more product they are making. Of course,” he adds, “the more complicated this system gets and the more enzymes you have, the more mutations you could make.” Because of that, he explains, a process like this would typically require hundreds and thousands of iterations (mutations), “and that is why it takes so long to engineer microorganisms.”

The software program can control aspects such as protein production across a 100,000-fold scale. “This is a very large dynamic range. It would be like if you had the volume on your stereo and the dial went from one to 100,000.”

The software will not only help companies that have already started on a project. Salis points out that companies starting from scratch can use the software to optimize a pathway, and in the lab his team has already performed more than 150 tests to prove the efficiency and accuracy of the software. The software is available free online. Established companies that want to test it out may want to try the version that can be licensed because, as Salis says, “if a company were to send a fancy enzyme to my Web server, I can’t guarantee it.” 

—Luke Geiver