GN-OA is a crystal structure prediction tool, which can predict crystal structures from scratch with extremely low computational cost.

  In GN-OA a graph network (GN) is adopted to establish a correlation model between the crystal structure and formation enthalpies, and optimization algorithm (OA) is used to accelerate the search for crystal structure with optimal formation enthalpy. The approach of combining GN and OA for crystal structure searching (GN-OA) can predict crystal structures at given chemical compositions with additional constraints on cell shapes and lattice symmetries.

System requirements


Getting started

1. Training the GN model

  We provided the origin MEGNet model in ./NN_model/, you just need to run this python file to get a trained GN model.

  Of course, you can also build a new GN model.

2. Crystal structure prediction


  If you use GN-OA for research, please consider citing this paper:
  Guanjian Cheng, Xin-Gao Gong, Wan-Jian Yin. “Crystal structure prediction by combining graph network and optimization algorithm.” Nat Commun 2022;13(1):1492. Doi: 10.1038/s41467-022-29241-4.