Researchers have used a popular AI model to redesign the enzyme behind Botox, creating more stable variants for directed evolution. The work, led by David Liu at the Broad Institute of Harvard and MIT, aims to overcome a key limitation: natural enzymes often collapse during directed evolution, a tedious and time-consuming process where scientists slowly nudge an enzyme toward new properties. “Laboratory evolution requires the commitment of time and resources. So what you start with is incredibly important as a major determinant of what you end up with,” said Liu in a press release. As a proof of concept, the redesigned enzymes were “far more stable and specific at cutting a protein linked to neurodegeneration compared to enzymes evolved from their natural counterparts.” Liu noted, “The most important finding is that using AI to stabilize natural proteins can provide much better starting points for laboratory protein evolution than what we and other researchers have been using for decades.” The strategy could expand the universe of designer enzymes, making it possible to target protein sequences currently out of reach because no suitable natural enzyme exists.