一致性规范图形神经网络用于分子性质预测
1Department of Industrial Engineering, Sungkyunkwan University, 2066 Seobu-ro Jangan-gu, Suwon, 16419, Republic of Korea.
概括
这项研究引入了一种一致性调节图形神经网络 (CRGNN),以改进对小数据集的分子性质预测. 该方法通过确保增强分子图的视图对齐来提高图形神经网络 (GNN) 的性能,克服了传统数据增强的局限性.
科学领域:
- 计算化学的计算化学
- 机器学习 机器学习
- 化学信息学 化学信息学
背景情况:
- 图形神经网络 (GNN) 在分子性质预测方面表现出色,但在小数据集方面却存在困难.
- 标准的数据增强技术往往会在分子图表中失败,可能会改变固有的特性.
研究的目的:
- 开发一种新的方法,一致性调节图形神经网络 (CRGNN),用于在GNN培训中有效的分子图形增强.
- 通过利用增强图形表示来增强小分子数据集上的GNN性能.
主要方法:
- 应用分子图增强,生成每个分子图的强增和弱增强视图.
- 引入了一致性正规化损失,以鼓励GNN在表示空间中密切地映射相同图的增强视图.
- 将此损失整合到GNN学习目标中.
主要成果:
- 该CRGNN方法有效地利用分子图增强,提高预测性能.
- 与现有方法相比,在各种分子基准数据集上表现出卓越的性能.
- 在较小的培训数据集上,性能增长尤其显著.
结论:
- 一致性规范化提供了一种可行的策略,以减轻GNN中分子图增强的负面影响.
- 拟议的CRGNN方法增强了对分子性质预测的数据增强的实用性,特别是在低数据模式中.
- 这种方法促进了GNN在小分子数据集的化学信息学中的应用.
更多相关视频
相关概念视频
Predicting Molecular Geometry
45.0K
VSEPR Theory for Determination of Electron Pair Geometries
45.0K
Network Covalent Solids
16.0K
Network covalent solids contain a three-dimensional network of covalently bonded atoms as found in the crystal structures of nonmetals like diamond, graphite, silicon, and some covalent compounds, such as silicon dioxide (sand) and silicon carbide (carborundum, the abrasive on sandpaper). Many minerals have networks of covalent bonds.
To break or to melt a covalent network solid, covalent bonds must be broken. Because covalent bonds are relatively strong, covalent network solids are typically...
To break or to melt a covalent network solid, covalent bonds must be broken. Because covalent bonds are relatively strong, covalent network solids are typically...
16.0K
Molecular Models
43.5K
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.
43.5K
Covalently Linked Protein Regulators
8.7K
Proteins can undergo many types of post-translational modifications, often in response to changes in their environment. These modifications play an important role in the function and stability of these proteins. Covalently linked molecules include functional groups, such as methyl, acetyl, and phosphate groups, and also small proteins, such as ubiquitin. There are around 200 different types of covalent regulators that have been identified.
These groups modify specific amino acids in a protein....
These groups modify specific amino acids in a protein....
8.7K
Covalently Linked Protein Regulators
2.0K
2.0K
Protein Networks
4.5K
An organism can have thousands of different proteins, and these proteins must cooperate to ensure the health of an organism. Proteins bind to other proteins and form complexes to carry out their functions. Many proteins interact with multiple other proteins creating a complex network of protein interactions.
These interactions can be represented through maps depicting protein-protein interaction networks, represented as nodes and edges. Nodes are circles that are representative of a protein,...
These interactions can be represented through maps depicting protein-protein interaction networks, represented as nodes and edges. Nodes are circles that are representative of a protein,...
4.5K


