Consider this: Polyethylene terephthalate (PET), a significant polymer found in most consumer packaging, accounts for 12% of global waste.

Engineers and scientists at the University of Texas, Austin, have used machine learning to create FAST-PETase, an enzyme variant that can break down plastic waste in a matter of hours or days at lower temperatures. Plastic waste is a serious environmental problem because it is generated in quantities of billions of tonnes, piles up in deadly landfills, pollutes land and oceans, and takes centuries to degrade. (UTNEWS)

How AI/ML helped create FAST-PETase

A plastic polymer known as polyethylene terephthalate (PET) makes up about 12% of global waste because it is used in the production of most consumer packaging.

A natural enzyme called PETase is known to allow bacteria to degrade PET plastics. Scientists at the Cockrell School of Engineering and College of Natural Sciences used machine learning to detect mutations of the enzyme that would allow for quick ‘depolymerization’ (breaking down into smaller constituents) of the plastic at low temperatures.

The AI model could identify an effective enzyme variant that has been named FAST-PETase (functional, active, stable, and tolerant PETase) by the researchers. After breaking down, the enzyme also ‘repolymerizes’ the plastic for reuse.

This is a significant breakthrough because so far it had not been possible to engineer enzymes for plastic recycling that were efficient at low temperatures, and were affordable and scalable for industrial application. FAST-PETase worked efficiently at less than 50 degrees Celsius.

This enzyme could potentially supercharge large-scale recycling of PET and allow industries to recover and reuse plastics at the molecular level.

“The possibilities are endless across industries to leverage this leading-edge recycling process,” said Hal Alper, professor in the McKetta Department of Chemical Engineering at UT Austin. “Beyond the obvious waste management industry, this also provides corporations from every sector the opportunity to take a lead in recycling their products. Through these more sustainable enzyme approaches, we can begin to envision a true circular plastics economy.”

“This work really demonstrates the power of bringing together different disciplines, from synthetic biology to chemical engineering to artificial intelligence,” said Andrew Ellington, professor in the Center for Systems and Synthetic Biology whose team led the development of the machine learning model.

The researchers have filed for a patent for their FAST-PETase technology.

Related Story: AI In The MARLIT APP To Keep Tabs On Plastic Marine Litter