Comparison Study on the Prediction of Multiple Molecular Properties by Various Neural Networks
- 4 October 2018
- journal article
- research article
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry A
- Vol. 122 (46) , 9128-9134
Abstract
Various neural networks, including single layer neural network (SLNN), deep neural network (DNN) with multi-layers and convolution neural network (CNN) have been developed and investigated to predict multiple molecular properties simultaneously. The data set of this work contains~134 kilo molecules and their 15 properties (including rotational constant A, B and C, dipole moment, isotropic polarizability, energy of HOMO, energy of LUMO, HOMO-LUMO gap energy, electronic spatial extent, zero point vibrational energy, internal energy at 0 K, internal energy at 298.15 K, enthalpy at 298.15 K, free energy at 298.15 K, and heat capacity at 298.15 K) at the hybrid density functional theory (DFT) level from the QM9 database. Coulomb matrix (CM) converted from the database representing every molecule uniquely and its eigenvalue are respectively used as the input of machine learning. The accuracies of predictions have been compared among SLNN, DNN and CNN by analyzing their mean absolute errors (MAEs). Using eigenvalues as input, both SLNN and DNN can give higher accuracy for the prediction of specific energy properties (U0, U, H and G). For the prediction of all 15 molecular properties at a time, DNN with 3-layers network exhibits the best results using the full CM as input. The number of layers in DNN play a key role in the prediction of multiple molecular properties simultaneously. This work may provide possibility and guidance for the selection of different neural networks and input data forms for prediction and validation of multiple parameters according to different needs.Keywords
Funding Information
- the national key Research and Development Program of China (2016YFB0600305)
- The National Natural Science Foundation of China (21306132)
This publication has 34 references indexed in Scilit:
- Assessment and Validation of Machine Learning Methods for Predicting Molecular Atomization EnergiesJournal of Chemical Theory and Computation, 2013
- Enumeration of 166 Billion Organic Small Molecules in the Chemical Universe Database GDB-17Journal of Chemical Information and Modeling, 2012
- Ruppet al.Reply:Physical Review Letters, 2012
- Prediction of Density and Viscosity of Biofuel Compounds Using Machine Learning MethodsEnergy & Fuels, 2012
- Extended-Connectivity FingerprintsJournal of Chemical Information and Modeling, 2010
- The calculation of thermodynamic properties of moleculesChemical Society Reviews, 2010
- Calculation of the standard molal thermodynamic properties as a function of temperature and pressure of some geochemically important organic sulfur compounds § §This paper is dedicated to Professor Harold C. Helgeson on the occasion of his seventieth birthday.Geochimica et Cosmochimica Acta, 2001
- Neural Networks in Computer IntelligenceTechnometrics, 1995
- Univariant Approximation by Superpositions of a Sigmoidal FunctionJournal of Mathematical Analysis and Applications, 1993
- Application of neural networks to computer recipe predictionColor Research & Application, 1991