Sign In

Communications of the ACM

ACM TechNews

Duke Software Dramatically Speeds Enzyme Design


View as: Print Mobile App Share: Send by email Share on reddit Share on StumbleUpon Share on Hacker News Share on Tweeter Share on Facebook

Duke University researchers have developed software that shows experimentalists how to alter the machinery that bacteria uses to make natural antibiotics. The program is a set of computer rules known as algorithm K* that can sort through all possible shapes and changes of a key enzyme that produces a natural antibiotic called gramicidin S, says Duke professor Bruce Donald. The new software could lead to more automated redesigning of old drugs to counter drug-resistant germs. "It is essentially a new pathway to make novel antibiotics," Donald says. "There are many possible changes you can make to a protein, but the algorithm can test out orders of magnitude more variations than laboratory experiments alone." He says the algorithm should enable researchers to quickly discover findings that would otherwise take longer through experimental techniques. "It should, in principle, be possible to redesign any enzyme simply by inputting the protein's shape into the algorithm and telling it what you want it to do," Donald says. The algorithm includes a "dead-end elimination" feature that can analyze all possible chemical interactions and flexible molecular architectures to eliminate scenarios that will not work. The K* algorithm is available as open source code for researchers to evaluate and use in their research.

Duke University
View Full Article

 

No entries found