AI allows algorithms to create biomolecules that have a wide range of useful functions

Shiran BarberZucker, a postdoctoral researcher at Prof. Sarel Fleisman’s lab, pursued an environmental goal: converting plastic waste into useful chemicals. Nature uses clever methods to decompose tough materials. For example, dead trees are recycled by white rot fungi whose enzymes convert wood into nutrients which return to soil. Why not use the same enzymes to degrade man-made wastes?

Barber-Zucker was troubled by the fact that these enzymes – called versatile peroxidases – are notoriously unstable. Fleishman of the Biomolecular Sciences Department of the Weizmann Institute of Science says, \”These natural enzymes can be real prima donnas. They are very difficult to work on.\” His lab developed computational methods used by thousands research teams worldwide to design enzymes with enhanced stability. To apply such methods, the molecular structure of a protein must be precisely known. This usually means that the proteins must be stable enough to form crystals which can then be bombarded by X-rays in order to reveal their 3D structure. The lab then uses its algorithms to tweak the structure and create a new protein that does not exist in nature.

Source:
https://phys.org/news/2022-04-ai-algorithms-biomolecules-huge-range.html