通过整合分子指纹和知识图特征进行基于神经网络的毒性预测
Junjie Xie1, Wei Liu1, Wei Hu1
1School of Informatics, Hunan University of Chinese Medicine, Changsha 410208, China.
Toxics
|November 27, 2025
概括
将知识图与图形神经网络 (GNN) 集成,可以改善分子毒性预测. 这种使用毒理学知识图 (ToxKG) 的新方法提高了药物查和风险评估的准确性和可解释性.
科学领域:
- 计算毒理学计算毒理学
- 化学信息学 化学信息学
- 生物信息学是一种生物信息学.
背景情况:
- 由于对生物机制的考虑有限,传统的分子毒性预测模型缺乏准确性和可解释性.
- 现有的方法通常仅依赖于分子结构特征,阻碍了概括性.
研究的目的:
- 开发一个新的分子毒性预测框架,整合知识图和图形神经网络 (GNN).
- 构建一个异质的毒理学知识图 (ToxKG),包含多种生物数据.
- 使用Tox21数据集评估各种GNN模型的性能.
主要方法:
- 使用ComptoxAI构建一个异质毒理学知识图 (ToxKG),整合PubChem,Reactome和ChEMBL的数据.
- 在Tox21数据集上对6个GNN模型 (GCN,GAT,R-GCN,HRAN,HGT,GPS) 的系统评估.
- 使用和不使用ToxKG丰富的GNN性能的比较.
主要成果:
- 用ToxKG信息丰富的异质图形模型显著优于传统的基于结构的模型.
- 包括AUC,F1得分,ACC和BAC在内的关键指标显示出了显著的改善.
- 在NR-AR受体任务中,GPS模型达到0.956的最高AUC,证明了生物机制的价值.
结论:
- 通过异质知识图和GNN集成生物机制对于准确的分子毒性预测至关重要.
- 这一框架为开发可解释和高效的智能毒理风险评估工具提供了一个有希望的方向.
更多相关视频
05:47Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems
Published on: June 13, 2025
1.3K
09:01A High-throughput Assay for the Prediction of Chemical Toxicity by Automated Phenotypic Profiling of Caenorhabditis elegans
Published on: March 14, 2019
7.6K
相关概念视频
01:36Types of Toxins
3.1K
Humans continually engage with an environment rich in potentially harmful chemicals. These are introduced to our bodies through inhalation, ingestion, or skin contact. These chemicals exist in various forms, such as air and environmental pollutants, agricultural chemicals, organic solvents, and heavy metals.
Air pollutants, primarily gases, pose significant threats to respiratory health, leading to conditions like hypoxia, lung cancer, and in extreme cases, death.
Environmental pollutants like...
Air pollutants, primarily gases, pose significant threats to respiratory health, leading to conditions like hypoxia, lung cancer, and in extreme cases, death.
Environmental pollutants like...
3.1K
02:26Protein 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
01:26Toxic Reactions: Overview
1.8K
When toxic substances penetrate the human body, they disseminate to various tissues, undergoing metabolic changes. This process yields reactive metabolites that may covalently bind with specific target molecules, resulting in toxicity.
Toxicity falls into two primary categories: local and systemic.
Local toxicity appears at the exposure site, such as protein denaturation caused by caustic substances.
In contrast, systemic toxicity requires the toxic agent's absorption and distribution,...
Toxicity falls into two primary categories: local and systemic.
Local toxicity appears at the exposure site, such as protein denaturation caused by caustic substances.
In contrast, systemic toxicity requires the toxic agent's absorption and distribution,...
1.8K