使用图形神经网络增强化学知识的有效学习分子性质
Tetiana Lutchyn1, Marie Mardal2,3, Benjamin Ricaud1
1Department of Physics and Technology, The Arctic University of Norway, Tromsø 9019, Norway.
ACS omega
|November 24, 2025
概括
将化学知识集成到图形神经网络 (GNN) 中,可以显著提高分子性质预测的准确性. 这种方法增强了GNN.
科学领域:
- 计算化学是一种计算化学.
- 机器学习 机器学习
- 药物发现 药物发现
背景情况:
- 图形神经网络 (GNN) 对于从结构数据中预测分子性质是有效的.
- 由于过度平滑和表达性挑战,GNN在捕捉全球分子特性方面面临限制.
- 现有的GNN模型难以内在地学习复杂的化学知识.
研究的目的:
- 开发基于GNN的模型,集成显式化学知识,以增强分子性质预测.
- 调查向GNN提供化学应用的全球图形信息的影响.
- 为了比较增强的GNN模型与纯GNN和大型基础模型的性能.
主要方法:
- 设计了一个简单的GNN架构,结合了特定领域的化学知识.
- 该模型在小分子数据集上进行训练,用于回归任务.
- 实现了节点级预测功能,以使用SMILES编码识别重要的分子亚结构.
- 性能被评估在几个基准与最先进的模型相比.
主要成果:
- 整合化学知识的GNN模型在准确性方面明显优于纯GNN方法.
- 改进后的模型表现出与更大,最先进的模型 (包括基础模型) 相比具有竞争力或更高的性能.
- 节点级预测使得能够识别影响预测的关键分子亚结构.
- 该模型通过适度的计算资源实现了高效的训练.
结论:
- 将化学知识整合到GNN中对于克服预测分子性质的局限性至关重要.
- 通过为GNN提供易于访问的全球图形信息,提高了它们在化学中的应用性.
- 开发的模型为分子性质预测提供了一个实用和准确的解决方案,适合广泛使用.
更多相关视频
相关概念视频
02:27Predicting Molecular Geometry
44.5K
VSEPR Theory for Determination of Electron Pair Geometries
44.5K
02:00Molecular Models
43.4K
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.
43.4K
01:27Inductive Effects on Chemical Shift: Overview
2.0K
The protons in unsubstituted alkanes are strongly shielded with chemical shifts below 1.8 ppm. Methine, methylene, and methyl protons appear at approximately 1.7, 1.2 and 0.7 ppm, while the proton signal from methane appears at 0.23 ppm. An electronegative substituent, such as chlorine, withdraws the electron density from the protons, increasing their chemical shift. Progressive substitution of the hydrogens in methane by chlorine shifts the proton signals increasingly downfield, to 3.05 ppm in...
2.0K
03:51Molecular Orbital Theory II
26.8K
Molecular Orbital Energy Diagrams
26.8K
01:26Chemical Synapses
11.1K
Chemical synapses are specialized sites between two neurons or between a neuron and a non-neuronal cell like a muscle, glandular or sensory cell.
Because chemical synapses depend on the release of neurotransmitter molecules from synaptic vesicles to pass on their signal, there is an approximately one millisecond delay between when the axon potential reaches the presynaptic terminal and when the neurotransmitter leads to opening of postsynaptic ion channels. Additionally, this signaling is...
Because chemical synapses depend on the release of neurotransmitter molecules from synaptic vesicles to pass on their signal, there is an approximately one millisecond delay between when the axon potential reaches the presynaptic terminal and when the neurotransmitter leads to opening of postsynaptic ion channels. Additionally, this signaling is...
11.1K
01:26Chemical Synapses
4.2K
Chemical synapses are specialized sites between two neurons or between a neuron and a non-neuronal cell like a muscle, glandular or sensory cell.
Because chemical synapses depend on the release of neurotransmitter molecules from synaptic vesicles to pass on their signal, there is an approximately one millisecond delay between when the axon potential reaches the presynaptic terminal and when the neurotransmitter leads to opening of postsynaptic ion channels. Additionally, this signaling is...
Because chemical synapses depend on the release of neurotransmitter molecules from synaptic vesicles to pass on their signal, there is an approximately one millisecond delay between when the axon potential reaches the presynaptic terminal and when the neurotransmitter leads to opening of postsynaptic ion channels. Additionally, this signaling is...
4.2K