Artificial intelligence can contribute to finding environmental solutions. Particularly in the field of recycling and decomposition of plastics. Engineers and scientists from the University of Texas at Austin have indeed created an enzyme with supernatural power. Developed using machine learning techniques, the lab-born protein is able to almost completely degrade PET packaging in just a week, a material that normally takes centuries to break down.
Reusing plastics at the molecular level
In their article published by the journal Nature, the researchers explain that their enzyme, called FAST-PETase, is also capable of carrying out a circular process consisting of breaking down the plastic into smaller elements (depolymerization), then to reconstitute it chemically (repolymerization) . All at a temperature of less than 50 degrees Celsius. “Until now, no one has been able to figure out how to make enzymes that can work efficiently at low temperatures to make them both portable and affordable on a large industrial scale,” says the University of Texas at Austin in its press release. . The substance created, which catalyzes chemical reactions, thus has the potential to boost large-scale recycling by recovering and reusing plastics at the molecular level. FAST-PETase also shows promise for environmental remediation. The research team is now studying a number of ways to clean up polluted sites using this artificial protein.
Five mutations of a natural enzyme
FAST-PETase is a mutant version (five mutations) of the naturally occurring enzyme PETase, also capable of degrading PET, but less rapidly. FAST-PETase was created using a machine learning algorithm that generated mutations of the natural enzyme and predicted those that would achieve the goal of rapid depolymerization at low temperatures. “This work really demonstrates the potential of bringing together different disciplines, from synthetic biology to chemical engineering to artificial intelligence,” said Andrew Ellington, a professor at the Center for Systems and Synthetic Biology, whose team led the development of the machine learning model.