An AI was tasked with creating proteins with anti-microbial properties. Researchers then created a subset of the proteins and found some did the job.
An AI has designed anti-microbial proteins that were then tested in real life and shown to work. The same approach could eventually be used to make new medicines.
Proteins are made of chains of amino acids. The sequence of those acids determine the protein’s shape and function.
Ali Madani at Salesforce Research in California and his colleagues used an AI to design millions of new proteins then created a small sample of those to test if they worked.
The AI, called ProGen, works in a similar way to AIs that can generate text. ProGen learned how to generate new proteins by learning “the grammar” of how amino acids combine from 280 million existing proteins. Instead of the team choosing a topic or style for the AI to write about, the researchers could specify a group of similar proteins for it to focus on.
Madani says that they programmed checks into the AI’s process so it would not produce amino acid “gibberish” but to really test it they wanted to see how the suggested molecules act in cells. Out of 100 that they physically created, 66 participated in similar chemical reactions to natural proteins that destroy bacteria in egg whites and saliva, suggesting they could also kill bacteria.
The researchers then imaged them with X-rays. Even though their amino acid sequences were up to 30 per cent different from any existing proteins, the shape they saw almost matched naturally occurring proteins. James Fraser at University of California San Francisco, who was part of the team, says it was not clear from the outset that the AI could work out how to change the amino acid sequence so much and still reproduced the correct shape.
“It was sort of an ‘it looks like a duck, it quacks like a duck’ situation and X-rays confirmed it also walked like a duck,” says Fraser. He was surprised to have found a well-functioning protein in the first relatively small fraction of all the ProGen-generated proteins that they tested.
A similar process could be used to create new test molecules for drug development, though they will still have to be tested in labs, which is time consuming, says Madani.