图形神经网络对分子几何学进行分类,并设计晶体和液体的新型顺序参数
Satoki Ishiai1, Katsuhiro Endo1, Kenji Yasuoka1
1Department of Mechanical Engineering, Keio University, Yokohama 223-8522, Japan.
The Journal of chemical physics
|August 8, 2023
概括
我们介绍TeaNet,一个用于分类分子结构的新型图形神经网络. 茶网使用最少的数据准确地识别结构,为分子分析提供了一种新的方法.
科学领域:
- 计算化学是一种计算化学.
- 材料科学 是一种材料科学.
- 机器学习 机器学习
背景情况:
- 分子动力学模拟产生复杂的3D结构数据.
- 分类分子结构传统上依赖于手动订单参数,这可能是劳动密集型的.
- 现有的机器学习方法往往需要大量数据和手动功能工程.
研究的目的:
- 开发一种自动化和高效的分类分子结构方法.
- 为了研究图形神经网络 (GNN) 在分子结构分类中的性能.
- 在数据要求和可解释性方面解决传统方法和深度学习模型的局限性.
主要方法:
- 应用两个GNN模型:图形卷积网络 (GCN) 和张量嵌入原子网络 (TeaNet).
- 这两种模型都被设计成不变于旋转和转移.
- 茶网利用原子和键之间的长度和方向信息来增强几何识别.
主要成果:
- 茶网通过非常小的分子结构 (分别为17和9个分子) 实现了高分类准确性 (LJ系统为98.9%,水系统为99.8%).
- 与需要更大的结构或更广泛的邻居信息的传统订单参数和GCN相比,TeaNet表现出优异的性能.
- 该研究证实了TeaNet能够自动生成新的可解释的顺序参数,将结构映射到一个低维特征空间.
结论:
- 茶网为分子结构分类提供了一种强大而高效的替代传统顺序参数.
- 它能够从小的局部结构中学习并产生可解释的特征,这使得它对结构分析有价值.
- 这项工作突出了先进的GNN在推进计算化学和材料科学方面的潜力.
相关概念视频
02:00Molecular Models
38.6K
Physical models representing molecular architectures of chemical compounds play essential roles in understanding chemistry. The use of molecular models makes it easier to visualize the structures and shapes of atoms and molecules.
38.6K
02:27Predicting Molecular Geometry
34.5K
VSEPR Theory for Determination of Electron Pair Geometries
34.5K
02:22Structures of Solids
14.3K
Solids in which the atoms, ions, or molecules are arranged in a definite repeating pattern are known as crystalline solids. Metals and ionic compounds typically form ordered, crystalline solids. A crystalline solid has a precise melting temperature because each atom or molecule of the same type is held in place with the same forces or energy. Amorphous solids or non-crystalline solids (or, sometimes, glasses) which lack an ordered internal structure and are randomly arranged. Substances that...
14.3K
01:18Molecular Shapes
57.0K
Molecules have characteristic shapes that are crucial for their function. The arrangement of various electron groups around the central atom dictates their molecular geometry. Electron pairs in the valence shell of a central atom will adopt an arrangement that minimizes repulsions between the electron pairs by maximizing the distance between them. The valence electrons form either bonding pairs, located primarily between bonded atoms, or lone pairs.
Two regions of electron density in a diatomic...
Two regions of electron density in a diatomic...
57.0K
01:21Polymer Classification: Crystallinity
2.9K
Unlike ionic or small covalent molecules, polymers do not form crystalline solids due to the diffusion limitations of their long-chain structures. However, polymers contain microscopic crystalline domains separated by amorphous domains.
Crystalline domains are the regions where polymer chains are aligned in an orderly manner and held together in proximity by intermolecular forces. For example, chains in the crystalline domains of polyethylene and nylon are bound together by van der Waals...
Crystalline domains are the regions where polymer chains are aligned in an orderly manner and held together in proximity by intermolecular forces. For example, chains in the crystalline domains of polyethylene and nylon are bound together by van der Waals...
2.9K
02:46Crystal Field Theory - Tetrahedral and Square Planar Complexes
42.9K
Tetrahedral Complexes
Crystal field theory (CFT) is applicable to molecules in geometries other than octahedral. In octahedral complexes, the lobes of the dx2−y2 and dz2 orbitals point directly at the ligands. For tetrahedral complexes, the d orbitals remain in place, but with only four ligands located between the axes. None of the orbitals points directly at the tetrahedral ligands. However, the dx2−y2 and dz2 orbitals (along the Cartesian axes) overlap with the ligands less than the dxy,...
Crystal field theory (CFT) is applicable to molecules in geometries other than octahedral. In octahedral complexes, the lobes of the dx2−y2 and dz2 orbitals point directly at the ligands. For tetrahedral complexes, the d orbitals remain in place, but with only four ligands located between the axes. None of the orbitals points directly at the tetrahedral ligands. However, the dx2−y2 and dz2 orbitals (along the Cartesian axes) overlap with the ligands less than the dxy,...
42.9K