文献类型：期刊文献

英文题名：Accurate predictions of aqueous solubility of drug molecules via the multilevel graph convolutional network (MGCN) and SchNet architectures

作者：Gao, Peng[1];Zhang, Jie[2,3];Sun, Yuzhu[3];Yu, Jianguo[3]

机构：[1]Univ Wollongong, Sch Chem & Mol Biosci, Wollongong, NSW 2500, Australia;[2]Guangzhou Regenerat Med & Hlth Guangdong Lab, Ctr Chem & Chem Biol, Bioland Lab, Guangzhou 53000, Peoples R China;[3]East China Univ Sci & Technol, Sch Chem Engn, Shanghai 200237, Peoples R China

年份：2020

卷号：22

期号：41

起止页码：23766

外文期刊名：PHYSICAL CHEMISTRY CHEMICAL PHYSICS

收录：;EI(收录号：20204709502970);WOS:【SCI-EXPANDED(收录号:WOS:000582937400027)】；

基金：We thank the Australian Government for supporting Peng's PhD study via providing him an Australian International Postgraduate Award scholarship. We also thank the NCI system, which is supported by the Australian Government (Project id: v15) for providing the computational resources to complete this project.

语种：英文

外文关键词：Forecasting - Network architecture - Solubility - Convolution - Deep learning - Drug delivery

摘要：Deep learning based methods have been widely applied to predict various kinds of molecular properties in the pharmaceutical industry with increasingly more success. In this study, we propose two novel models for aqueous solubility predictions, based on the Multilevel Graph Convolutional Network (MGCN) and SchNet architectures, respectively. The advantage of the MGCN lies in the fact that it could extract the graph features of the target molecules directly from the (3D) structural information; therefore, it doesn't need to rely on a lot of intra-molecular descriptors to learn the features, which are of significance for accurate predictions of the molecular properties. The SchNet performs well in modelling the interatomic interactions inside a molecule, and such a deep learning architecture is also capable of extracting structural information and further predicting the related properties. The actual accuracy of these two novel approaches was systematically benchmarked with four different independent datasets. We found that both the MGCN and SchNet models performed well for aqueous solubility predictions. In the future, we believe such promising predictive models will be applicable to enhancing the efficiency of the screening, crystallization and delivery of drug molecules, essentially as a useful tool to promote the development of molecular pharmaceutics.