Researchers at the University of Texas at Austin used a machine learning tool to predict which mutations would make a plastic-eating enzyme more stable, producing a variant named FAST-PETase, tested in vitro on real plastic waste rather than in any animal or living organism.
FAST-PETase almost completely broke down untreated plastic from 51 different consumer products within a week at mild temperatures, and it also broke down an entire heat-treated water bottle, outperforming earlier engineered versions of the same enzyme.
The enzyme still needs testing at an industrial scale before it changes how plastic actually gets recycled, and so far the results come from lab-scale batches of collected waste, not a working recycling plant. The new enzyme differs from the natural one by five amino acid changes, three of them picked by the model.
Read the original study
Lu et al., Nature, 2022 · doi.org/10.1038/s41586-022-04599-z
Software picked the changes, and the enzyme turned fast.
Source
- Lu, H., Diaz, D. J., Czarnecki, N. J., Zhu, C., Kim, W., Shroff, R., Acosta, D. J., Alexander, B. R., Cole, H. O., Zhang, Y., Lynd, N. A., Ellington, A. D., & Alper, H. S. (2022). Machine learning-aided engineering of hydrolases for PET depolymerization. Nature, 604(7907), 662–667. https://doi.org/10.1038/s41586-022-04599-z