Geometric graph neural networks (GNNs) have emerged as powerful tools for modeling molecular geometry. However, they encounter limitations in effectively capturing long-range interactions in large molecular systems. To address this challenge, we introduce Neural P3M, a versatile enhancer of geometric GNNs to expand the scope of their capabilities by incorporating mesh points alongside atoms and reimaging traditional mathematical operations in a trainable manner. Neural P3M exhibits flexibility across a wide range of molecular systems and demonstrates remarkable accuracy in predicting energies and forces, outperforming on benchmarks such as the MD22 dataset. It also achieves an average improvement of 22% on the OE62 dataset while integrating with various architectures.
@article{arxiv.2409.17622,
title = {Neural P$^3$M: A Long-Range Interaction Modeling Enhancer for Geometric GNNs},
author = {Yusong Wang and Chaoran Cheng and Shaoning Li and Yuxuan Ren and Bin Shao and Ge Liu and Pheng-Ann Heng and Nanning Zheng},
journal= {arXiv preprint arXiv:2409.17622},
year = {2024}
}