Molecular Property Prediction: A Multilevel Quantum Interactions Modeling Perspective
Authors:
Open Access
- 17 July 2019
- journal article
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 33 (01) , 1052-1060
Abstract
Predicting molecular properties (e.g., atomization energy) is an essential issue in quantum chemistry, which could speed up much research progress, such as drug designing and substance discovery. Traditional studies based on density functional theory (DFT) in physics are proved to be time-consuming for predicting large number of molecules. Recently, the machine learning methods, which consider much rule-based information, have also shown potentials for this issue. However, the complex inherent quantum interactions of molecules are still largely underexplored by existing solutions. In this paper, we propose a generalizable and transferable Multilevel Graph Convolutional neural Network (MGCN) for molecular property prediction. Specifically, we represent each molecule as a graph to preserve its internal structure. Moreover, the well-designed hierarchical graph neural network directly extracts features from the conformation and spatial information followed by the multilevel interactions. As a consequence, the multilevel overall representations can be utilized to make the prediction. Extensive experiments on both datasets of equilibrium and off-equilibrium molecules demonstrate the effectiveness of our model. Furthermore, the detailed results also prove that MGCN is generalizable and transferable for the prediction.Keywords
All Related Versions
This publication has 20 references indexed in Scilit:
- Machine Learning of Dynamic Electron Correlation Energies from Topological AtomsJournal of Chemical Theory and Computation, 2017
- Prediction Errors of Molecular Machine Learning Models Lower than Hybrid DFT ErrorJournal of Chemical Theory and Computation, 2017
- Representations in neural network based empirical potentialsThe Journal of Chemical Physics, 2017
- Quantum-chemical insights from deep tensor neural networksNature Communications, 2017
- Molecular graph convolutions: moving beyond fingerprintsJournal of Computer-Aided Molecular Design, 2016
- Machine Learning Predictions of Molecular Properties: Accurate Many-Body Potentials and Nonlocality in Chemical SpaceThe Journal of Physical Chemistry Letters, 2015
- Enumeration of 166 Billion Organic Small Molecules in the Chemical Universe Database GDB-17Journal of Chemical Information and Modeling, 2012
- Application of molecular dynamics simulations in molecular property prediction II: Diffusion coefficientJournal of Computational Chemistry, 2011
- A new hybrid exchange–correlation functional using the Coulomb-attenuating method (CAM-B3LYP)Chemical Physics Letters, 2004
- Theory of complex molecular interactions: computer graphics, distance geometry, molecular mechanics, and quantum mechanicsAccounts of Chemical Research, 1985