使用机器学习预测有机化合物的水溶性:对基于描述器和基于指纹的模型进行比较研究
Arash Tayyebi1, Ali S Alshami2, Zeinab Rabiei3
1University of North Dakota, Chemical Engineering, Grand Forks, ND, 58201, USA.
机器学习模型使用分子描述符和指纹准确地预测化学水溶性. 物理化学描述器实现了更高的预测准确性,为材料设计提供了具有成本效益的工具.
科学领域:
- 计算化学计算化学
- 材料科学 材料科学 材料科学
- 机器学习 机器学习
背景情况:
- 确定化学水溶性的传统方法依赖于经验观测和广泛的实验.
- 可溶性的预测模型可以加速新材料和化学配方的开发.
- 机器学习 (ML) 提供了一种数据驱动的方法来预测化学性质.
研究的目的:
- 开发和评估用于预测化学物种水溶性的机器学习模型.
- 为了比较分子描述符和摩根指纹的性能,作为可溶性预测的输入特征.
- 使用可解释性技术,识别影响水溶性的关键特征.
主要方法:
- 两种机器学习模型使用随机森林 (RF) 训练了超过8400个化合物的数据集.
- 分子描述符和摩根指纹被用作特征集来表示化学结构.
- 模型性能使用确定系数 (R2) 和根-平均-平方偏差 (RMSE) 进行评估,可通过沙普利增量解释 (SHAP) 进行解释.
主要成果:
- 与指纹模型 (R2 = 0.81,RMSE = 0.80) 相比,物理化学描述模型实现了较高的R2为0.88和较低的RMSE为0.64.
- 指纹方法为分子相互作用和热力学兼容性提供了洞察力.
- SHAP分析确定了有助于水溶性预测的关键特征.
结论:
- 机器学习模型,特别是那些使用物理化学描述符的机器学习模型,提供了对水溶性的准确和高效预测.
- 这些数据驱动工具可以显著降低与材料开发相关的成本和时间.
- 基于描述器的模型显示了对测试数据集的卓越预测准确性.
更多相关视频
09:21Author Spotlight: Generating Neuronal Phenotypic Profiles - A Protocol to Culture and Image Human Midbrain Dopaminergic Neurons
Published on: July 7, 2023
10:25Construction of Models for Nondestructive Prediction of Ingredient Contents in Blueberries by Near-infrared Spectroscopy Based on HPLC Measurements
Published on: June 28, 2016
相关概念视频
Pharmacokinetic Models: Comparison and Selection Criterion
Physiological models take a detailed approach by considering specific molecular processes. They can predict drug distribution, metabolism, and elimination changes, providing a comprehensive understanding of how drugs interact with the body.
Entropy and Solvation
Factors Affecting Solubility
Physical Properties Affecting Solubility
As for any solution, the solubility of a gas in a liquid is affected by the attractive intermolecular forces between solute and solvent species. Unlike solid and liquid solutes, however, there is no solute-solute intermolecular attraction to overcome when a gaseous solute dissolves in a liquid solvent since the atoms or molecules comprising a gas are far separated and experience negligible interactions. Consequently, solute-solvent interactions are the sole...
Classification of Titrimetric Analysis Based on Reaction Types
Titrations between an acid and a base lead to neutralization reactions that form...
Mechanistic Models: Compartment Models in Individual and Population Analysis
