Google AI
“AlphaFold 3” predicts the structure of all molecules of life
The model from the AlphaFold Protein Structure Database represents the backbone of the protein structure. So-called secondary structure elements are represented as bands. In the blue areas the model is probably reliable. The yellow areas are likely flexible, and only one possible structure is shown. photo
© -/AlphaFold Protein Structure Database/dpa
Until now, it has been tedious to find out exactly what proteins do. But that is changing rapidly with the help of artificial intelligence. AI is expected to achieve a breakthrough in molecular research.
The DeepMind subsidiary Isomorphic Labs from London, which specializes in the use of artificial intelligence in drug research, also played a key role in the development of the AI model. DeepMind had previously developed “AlphaFold 2,” a program that can calculate the three-dimensional shape of proteins.
Malaria vaccines, cancer treatments and enzyme design
Researchers assume that biological processes in the body and diseases can be better understood if the shape and thus the function of proteins are known. In 2020, DeepMind achieved a fundamental breakthrough in protein structure prediction. So far, millions of researchers worldwide have used AlphaFold 2 to make discoveries in areas such as malaria vaccines, cancer treatments and enzyme design.
In an article published in the journal Nature, DeepMind presented the AI system “AlphaFold 3” as a model that could predict the structure and interaction of all molecules of life with unprecedented accuracy. “In the interactions of proteins with other types of molecules, we see an improvement of at least 50 percent compared to existing prediction methods, and in some important categories of interactions we have doubled the prediction accuracy.”
Google DeepMind also announced the “AlphaFold Server.” This free tool gives non-commercial researchers access to the capabilities of “AlphaFold 3” to generate models of biological structures. It enables biologists to create large and complex protein structures with just a few mouse clicks on the screen. In order to exploit the potential of “AlphaFold 3” for drug development, Isomorphic Labs is already working with pharmaceutical companies.