In 2021, AI research laboratory Deep mind announced the development of its first digital biology neural network, AlphaFold. The model was able to accurately predict the three-dimensional structure of proteins, which determines the functions these molecules play. “We are simply floating bags of water that move,” says Pushmeet Kohli, vice president of research at DeepMind. “What makes us special are proteins, the building blocks of life. “The way they interact with each other is what makes the magic of life happen.”
AlphaFold was considered by Science magazine as the breakthrough of the year in 2021. In 2022, it was the The most cited research article in AI. “People have been on (protein structures) for many decades and couldn’t make that much progress,” Kohli says. “Then came AI.” DeepMind also launched the AlphaFold Protein Structure Database—which contained the protein structures of almost every organism whose genome has been sequenced—making it available to scientists around the world for free.
More than 1.7 million researchers in 190 countries They have used it for research ranging from designing enzymes that eat plastic to developing more effective malaria vaccines. A quarter of the research involving AlphaFold was dedicated to understanding cancer, Covid-19 and neurodegenerative diseases such as Parkinson’s and Alzheimer’s. Last year, DeepMind launched its next generation of AlphaFold, which extended its structure prediction algorithm to biomolecules such as nucleic acids and ligands.
“It has democratized scientific research,” Kohli says. “Scientists working in a developing country on a neglected tropical disease did not have access to the funds to calculate the structure of a protein. Now, with the click of a button, they can go to the AlphaFold database and get these predictions for free.” For example, one of DeepMind’s first partners, the Drugs for Neglected Diseases Initiative, used AlphaFold to develop drugs for diseases that affect millions of people (such as sleeping sickness, Chagas disease, and leishmaniasis) and that, however, they receive comparatively little research.
DeepMind’s latest advancement is called AlphaMissense. The model classifies so-called missense mutations: genetic alterations that can result in the production of different amino acids at particular positions in proteins. Such mutations can alter the function of the protein itself, and AlphaMissense attributes a probability score to whether that mutation is pathogenic or benign. “Understanding and predicting these effects is crucial for the discovery of rare genetic diseases,” says Kohli. The algorithm, which was launched last year, has classified about 89 percent of all possible human errors. Until now, researchers had only clinically classified 0.1 percent of all possible variants.
“This is just the beginning,” says Kohli. Ultimately, he believes AI could lead to the creation of a virtual cell that could radically accelerate biomedical research, allowing biology to be explored in-silico rather than in real-world laboratories. “With AI and machine learning we finally have the tools to understand this very sophisticated system we call life.”
This article appears in the July/August 2024 issue of UK WIRED Magazine.