通过稀疏学习构建可解释图形神经网络,用于药物-蛋白质结合预测
Yang Wang1, Zanyu Shi2, Pathum Weerawarna3
1Computer Science Department, Luddy School of Informatics, Computing, and Engineering, Indiana University Bloomington, Bloomington, Indiana, USA.
精细学习图形神经网络 (SLGNN) 识别化学有效的药物结构,用于蛋白质结合的预测. 这种方法克服了当前可解释的GNN模型的局限性,提高了准确性和可解释性.
科学领域:
- 计算化学计算化学
- 生物信息学是一种生物信息学.
- 机器学习 机器学习
背景情况:
- 可解释图形神经网络 (GNN) 用于药物与蛋白质结合的预测.
- 当前的GNN模型经常识别出化学无效的关键结构.
- 需要手动设置值以精确确定重要的基层结构.
研究的目的:
- 开发一种新的可解释的GNN模型来预测药物与蛋白质的结合.
- 确保已识别的关键药物结构具有化学有效性.
- 提高基于GNN的药物向相互作用分析的准确性和可解释性.
主要方法:
- 拟议的稀疏学习图表神经网络 (SLGNN).
- 使用基于化学子结构的药物分子图表表示.
- 整合了通用化的合拉索与消息传递算法.
主要成果:
- SLGNN成功地确定了化学有效的基结构,这些基结构对于药物与蛋白质的结合至关重要.
- 与最先进的方法相比,所识别的子结构显示出更强的预测能力.
- SLGNN 消除了手动值选择的需要.
结论:
- SLGNN为药物蛋白结合预测提供了一个强大的和可解释的方法.
- 该方法确保了已识别的关键药物结构的化学有效性.
- SLGNN在药物发现和开发中推进了可解释的AI.
更多相关视频
08:31Biosensor-based High Throughput Biopanning and Bioinformatics Analysis Strategy for the Global Validation of Drug-protein Interactions
Published on: December 1, 2020
06:50Author Spotlight: A Computational Approach to Decipher Amino Acid Preferences in Multispecific Protein-Protein Interactions
Published on: January 26, 2024
相关概念视频
Protein-Drug Binding: Determination Methods
Indirect methods involve isolating the bound drug from its free form in biological samples such as blood, serum, or plasma. These techniques aim to measure the percentage of drugs bound to proteins. Equilibrium dialysis is a commonly used method where the free drug concentration at equilibrium is measured by separating the bound...
Drug Distribution: Plasma Protein Binding
Protein-Drug Binding: Mechanism and Kinetics
Various forces drive these interactions, including hydrogen bonds, hydrophobic interactions, ionic bonds, electrostatic interactions, and van der Waals forces. These bonds enable drugs to bind to specific sites on proteins,...
Protein Networks
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,...
Protein-protein Interfaces
Physiological Pharmacokinetic Models: Assumption with Protein Binding
