Protein shells designed using AI had no imperfections
Shutterstock/Sergei Drozd
AI can engineer extremely dense protein shells that could one day lead to more potent vaccines.
The genetic material of viruses is housed in protein shells. Similar lab-made shells are used in vaccines, encapsulating molecules that induce an immune response in the body. The chemical and biological properties of these artificially made shells depend on their construction – any imperfection, no matter how small, makes them less effective, making them unstable and reacting in unpredictable ways inside the cells. Isaac Lutz of the University of Washington in Seattle and his colleagues wanted to see if the use of artificial intelligence could make the design and creation of these shells more precise.
They first provided the AI with the properties they wanted a shell to have, such as its size and porosity. The AI then used reinforcement learning – the same iterative process that AI systems use to learn how to play games like chess by trying different moves, then receiving feedback and trying again. Here, the AI’s moves consisted of combining, folding, and intertwining small protein structures called alpha helices into 20-sided shells, and then checking whether the design had the desired properties.
After the AI designed hundreds of thousands of shells, the researchers created around 350 in the lab. They examined them under an electron microscope and found that the AI had created denser shells than ever before. Lutz says that’s because he started with very small building blocks that could be made to fit together more perfectly than the larger protein structures the researchers used before.
“It’s like before we had to buy something like protein Legos, and what you could build was limited by how well they could fit together. Whereas now we can say what we want to build, the AI designs and connects exactly the Legos needed to achieve this,” he says.
To test how the high density affects the shells’ uses in living cells, the team studded the shells with different molecules and inserted them into mice. Notably, in an experiment with molecules that trigger the production of influenza antibodies, the AI-designed shells resulted in a small but statistically significant increase in immune response compared to some more conventional vaccine candidates that are currently in clinical trials. . Lutz says this is due to the precision of the AI method – each molecule is exactly where it needs to be on the shell, and the shell is structurally strong enough to support several of them.
“It’s amazing that the team could do that. It takes billions of years of evolution to design unique proteins that fold up perfectly, but that’s another level of complexity, to fold up proteins to fit together so well and form structures closed,” says Martin Noble of the University of Newcastle in the UK.
Yang Zhang of the University of Michigan says that in addition to vaccines, AI-designed protein shells could be useful for gene therapy where genetic material could be placed inside a shell designed for patient’s cells do not react negatively to it.
Topics:
