PI: Bruce Tidor, Department of Electrical Engineering & Computer Science, MIT
Enzymes are proteins that, as catalysts, speed and tune key chemical reactions in nature. Industries also use enzymes––those from nature as well as those engineered by humans––in a wide range of applications, from the production of pharmaceuticals, fine chemicals, biofuels, and other goods to the cleanup of environmental contamination, and more. An important limitation in many industrial processes is the efficient engineering of needed enzymatic function. This study applies computer modeling and machine learning approaches, in tandem, to better understand what features help to make an efficient enzyme and as tools for engineering improved enzymes with desired properties. Results of this work hold the potential to dramatically improve the implementation speed and effectiveness of many industrial processes.