文献类型：期刊文献

英文题名：A general QSPR protocol for the prediction of atomic/inter-atomic properties: a fragment based graph convolutional neural network (F-GCN)

作者：Gao, Peng[1];Zhang, Jie[2,3];Qiu, Hongbo[4];Zhao, Shuaifei[5]

机构：[1]Univ Wollongong, Sch Chem & Mol Biosci, Wollongong, NSW 2500, Australia;[2]Ctr Chem & Chem Biol, Bioland Lab, Guangzhou Regenerat Med & Hlth Guangdong Lab, Guangzhou 53000, Peoples R China;[3]East China Univ Sci & Technol, Sch Chem Engn, Shanghai 200237, Peoples R China;[4]Monash Univ, Dept Chem Engn, Clayton, Vic 3800, Australia;[5]Deakin Univ, Inst Frontier Mat IFM, Perth, WA, Australia

年份：2021

卷号：23

期号：23

起止页码：13242

外文期刊名：PHYSICAL CHEMISTRY CHEMICAL PHYSICS

收录：;EI(收录号：20212510541841);WOS:【SCI-EXPANDED(收录号:WOS:000657870200001)】；

基金：We acknowledge the NCI system, supported by the Australian Government (Project id: v15) for providing computational resources. We thank the Australian Government, which offered P. G. an Australian International Postgraduate Award scholarship to complete his PhD study.

语种：英文

外文关键词：Convolution - Density functional theory - Graph neural networks - Chemical bonds - Atoms - Computational chemistry - Design for testability - Computation theory - Convolutional neural networks

摘要：In this study, a general quantitative structure-property relationship (QSPR) protocol, fragment based graph convolutional neural network (F-GCN), was developed for the prediction of atomic/inter-atomic properties. We applied this novel artificial intelligence (AI) tool in predictions of NMR chemical shifts and bond dissociation energies (BDEs). The obtained results were comparable to experimental measurements, while the computational cost was substantially reduced, with respect to pure density functional theory (DFT) calculations. The two important features of F-GCN can be summarised as: first, it could utilise different levels of molecular fragments for atomic/inter-atomic information extraction; second, the designed architecture is also open to include additional descriptors for a more accurate solution of the local environment at atomic level, making itself more efficient for structural solutions. And during our test, the averaged prediction error of H-1 NMR chemical shifts is as small as 0.32 ppm, and the error of C-H BDE estimation is 2.7 kcal mol(-1). Moreover, we further demonstrated the applicability of this developed F-GCN model via several challenging structural assignments. The success of the F-GCN in atomic and inter-atomic predictions also indicates an essential improvement of computational chemistry with the assistance of AI tools.