Chinese scientists have developed the world’s first drug-molecule property prediction system using a quantum-embedded graph neural network, according to Science and Technology Daily. Created by Origin Quantum with partners at the University of Science and Technology of China and the Hefei Comprehensive National Science Center, the system can simultaneously process atoms and chemical bonds at the quantum level, overcoming a limitation in existing quantum algorithms. This advance greatly improves the accuracy of molecular behavior predictions, boosting drug discovery efficiency. Tests on the Origin Wukong quantum computer confirmed the model’s stable performance despite current quantum hardware limitations, and the findings were published in the Journal of Chemical Information and Modeling.
Credit: CGTN