HetDDI:一种预训练的异构图神经网络模型,用于药物相互作用预测
Zhe Li1, Xinyi Tu1, Yuping Chen2
1School of Computer Science, University of South China, Hengyang, 421001 Hunan, China.
Briefings in bioinformatics
|October 30, 2023
概括
预测药物相互作用 (DDI) 由于日益增长的多药性,至关重要. 一种新的预训练异质图形神经网络 (HGNN) 模型,HetDDI,有效地整合了药物结构和生物医学知识,以准确预测DDI.
科学领域:
- 生物信息学是一种生物信息学.
- 计算药理学计算药理学
- 药物发现 药物发现 药物发现
背景情况:
- 增加多种疾病的并发症需要同时使用药物,从而增加不良药物相互作用的风险.
- 准确的DDI预测对于临床安全和生物信息学研究至关重要.
- 现有的方法可能无法充分利用药物结构信息和外部生物医学知识.
研究的目的:
- 开发一种新的预训练异质图形神经网络 (HGNN) 模型,HetDDI,用于预测药物相互作用.
- 有效地整合药物分子结构信息与生物医学知识图表中的丰富语义信息.
- 提高DDI预测在各种任务和数据集中的准确性和可靠性.
主要方法:
- 提出 HetDDI,一个预训练的 HGNN 模型.
- 使用各种预训练策略初始化模型参数.
- 学习药物特征表示通过聚合多源异质信息,包括药物结构和生物医学知识.
- 在三个数据集和三个场景 (S1,S2,S3) 中对二进制类,多类和多标签DDI预测任务的评估性能.
主要成果:
- 在S1.上实现了高精度:98.82% (二进制类),98.13% (多类) 和96.66% (多标签).
- 在S1.1上,超越最先进的方法的性能至少为2%.
- 在S2和S3情景中表现出强的表现.
- 案例研究证实了该模型在预测新型DDI方面的有效性.
结论:
- 预训练的HetDDI模型有效地利用了固有的药物特性和外部的生物医学知识,以进行卓越的DDI预测.
- HetDDI在预测药物相互作用方面取得了重大进展,有助于提高患者的安全性.
- 该模型在多个任务和数据集中的表现突显了其稳定性和潜在的临床适用性.
相关概念视频
Quantitative Aspects of Drug-Receptor Interaction
1.0K
The receptor occupancy theory connects a drug's response to the number of occupied receptors. With higher drug concentrations, more receptors are occupied, leading to increased responses. The formation of drug-receptor complexes involves association and dissociation rates, which reach equilibrium when the forward and backward reactions are equal. The equilibrium association constant (Ka) and its inverse, the equilibrium dissociation constant (Kd), indicate drug affinity. Higher Ka and lower...
1.0K
Drug-Receptor Interactions
5.3K
Drug-receptor interaction describes the binding of receptors by drugs, but not all drug-receptor interactions result in activation and tissue response. For instance, the binding of agonists activates the receptor to generate a cellular reaction, while antagonists bind to receptors without causing their activation.
Several parameters, such as the drug's affinity for its receptor and its efficacy, which is its ability to activate the receptor, determine the drug's effect on the tissue....
Several parameters, such as the drug's affinity for its receptor and its efficacy, which is its ability to activate the receptor, determine the drug's effect on the tissue....
5.3K
Protein-protein Interfaces
12.5K
Many proteins form complexes to carry out their functions, making protein-protein interactions (PPIs) essential for an organism's survival. Most PPIs are stabilized by numerous weak noncovalent chemical forces. The physical shape of the interfaces determines the way two proteins interact. Many globular proteins have closely-matching shapes on their surfaces, which form a large number of weak bonds. Additionally, many PPIs occur between two helices or between a surface cleft and a...
12.5K
Drug-Receptor Interaction: Antagonist
3.0K
An antagonist is a drug that binds strongly to a receptor without activating it. An antagonist prevents other molecules, such as neurotransmitters or hormones, from binding to the receptor and triggering a cellular response. Such interaction effectively hinders the normal physiological processes mediated by the receptor, resulting in various pharmacological effects depending on the specific receptor targeted.
Antagonists can be classified as competitive or noncompetitive based on their...
Antagonists can be classified as competitive or noncompetitive based on their...
3.0K
Combined Effects of Drugs: Synergism
4.0K
Synergism is a useful mechanism where combining two or more drugs is more effective than each constituent used alone. Such combinations are also called supra-additive interactions. The drugs collectively enhance the final therapeutic effect by acting on different targets. Another advantage is that the low dose of each constituent drug is sufficient to achieve the desired effect. This helps reduce the duration of therapy and lower the adverse effects of these drugs.
Such synergistic combinations...
Such synergistic combinations...
4.0K
Model Approaches for Pharmacokinetic Data: Distributed Parameter Models
74
Pharmacokinetic models are mathematical constructs that represent and predict the time course of drug concentrations in the body, providing meaningful pharmacokinetic parameters. These models are categorized into compartment, physiological, and distributed parameter models.
The distributed parameter models are specifically designed to account for variations and differences in some drug classes. This model is particularly useful for assessing regional concentrations of anticancer or...
The distributed parameter models are specifically designed to account for variations and differences in some drug classes. This model is particularly useful for assessing regional concentrations of anticancer or...
74


