The AlphaFold of materials invented millions of new materials that have not yet been synthesized
source：PUHUA system 2023-08-01
University of California San Diego nanoengineering professor Shyue Ping Ong(Pymatgen's team) describes M3GNet as "the AlphaFold of materials" (AlphaFold is a breakthrough AI algorithm built by Google's DeepMind that can predict protein structures).
Professor Ong said: "The design development of new materials is similar to that of proteins, we need to understand the structure of a material to predict its properties. We strongly believe that the M3GNet architecture is a transformative tool that can greatly expand our ability to explore the chemistry and structure of new materials."
The Matterver.ai database currently contains more than 31 million new materials that have not yet been synthesized, of which more than 1.8 million are stable enough to be used for further applications.
Reference:
1. Chen, C., Ong, S.P. A universal graph deep learning interatomic potential for the periodic table. Nat Comput Sci 2, 718–728 (2022). https://doi.org/10.1038/s43588-022-00349-3
2. M3GNet: https://github.com/materialsvirtuallab/m3gnet
3. Matterverse.AI Database: http://matterverse.ai
