A team of US researchers has developed an AI system that can produce synthetic enzymes from scratch. Even though several of these enzymes' artificially created amino acid sequences differed noticeably from those of any known natural protein, they still performed as well in laboratory testing as in nature.

The experiment shows that, despite being designed to read and create English text, natural language processing can pick up on at least some of the fundamental concepts of biology. For example, the artificial intelligence (AI) tool ProGen, created by Salesforce Research, assembles amino acid sequences into synthetic proteins using next-token prediction.

According to scientists, the new technology could surpass directed evolution, a protein design technique that won a Nobel Prize. It will revitalise the 50-year-old field of protein engineering by accelerating the creation of novel proteins that can be used for anything, from therapeutics to degrading plastic.

Furthermore, the AI could determine how the enzymes should be shaped by analysing the raw sequence data. As defined by X-ray crystallography, the manufactured proteins' atomic structures seemed just as they should, although their sequences were novel.