A new geometric machine learning method promises to accelerate precision drug development
Peer-Reviewed Publication
Updates every hour. Last Updated: 11-Jul-2025 02:10 ET (11-Jul-2025 06:10 GMT/UTC)
Proteins play a crucial role in nearly all biological processes, yet predicting their complex interactions and designing proteins with new functions poses a significant challenge. In a new study published in Nature (DOI: 10.1038/s41586-024-08435-4), researchers at EPFL in Lausanne and AITHYRA of the Austrian Academy of Sciences (ÖAW) in Vienna have successfully used AI techniques to computationally predict the newly formed “neo-surfaces” of proteins after small drug molecules bind to them and design artificial proteins that can bind to these new surfaces. The computational results were confirmed experimentally in vitro. The new method has the potential to accelerate the development of future precision drugs.
Experts are warning of the risks of spreading invasive and non-native species when moving large volumes of untreated lake, reservoir and river water.
New research from the University of Chicago revealed that the nervous system circuitry that controls arm movement in octopuses is segmented, giving these extraordinary creatures precise control across all eight arms and hundreds of suckers to explore their environment, grasp objects, and capture prey.