AI Transforms Protein Engineering with Artificial Enzyme Generation

Protein engineering is about to be transformed with the development of ProGen, an AI system capable of generating artificial enzymes from scratch. Salesforce Research used natural language processing to teach the machine the underlying principles of biology, leading to the generation of amino acid sequences into artificial proteins.

In vitro tests have shown that some of the artificial enzymes generated by ProGen have activity comparable to naturally occurring proteins, even when their sequences differ significantly. This breakthrough has the potential to revolutionize the field of protein engineering and lead to new therapeutic applications.

Results of the In Vitro Tests

Out of the first batch of 100 proteins, five artificial proteins were tested in cells and compared to the activity of an enzyme found in the whites of chicken eggs. Two of the artificial enzymes were able to break down the cell walls of bacteria with activity comparable to the natural enzyme, yet their sequences were only about 18% identical. The AI was even able to learn how the enzymes should be shaped, simply from studying the raw sequence data. Measured with X-ray crystallography, the atomic structures of the artificial proteins looked just as they should, although the sequences were like nothing seen before. The AI-generated enzymes showed activity even when as little as 31.4% of their sequence resembled any known natural protein.

The future looks bright for protein engineering and the possibilities are endless with this new tool at our disposal.