Improving the accuracy of Møller-Plesset perturbation theory with neural networks
Open Access
- 6 September 2017
- journal article
- conference paper
- Published by AIP Publishing in The Journal of Chemical Physics
- Vol. 147 (16) , 161725
Abstract
Noncovalent interactions are of fundamental importance across the disciplines of chemistry, materials science, and biology. Quantum chemical calculations on noncovalently bound complexes, which allow for the quantification of properties such as binding energies and geometries, play an essential role in advancing our understanding of, and building models for, a vast array of complex processes involving molecular association or self-assembly. Because of its relatively modest computational cost, second-order Møller-Plesset perturbation (MP2) theory is one of the most widely used methods in quantum chemistry for studying noncovalent interactions. MP2 is, however, plagued by serious errors due to its incomplete treatment of electron correlation, especially when modeling van der Waals interactions and π-stacked complexes. Here we present spin-network-scaled MP2 (SNS-MP2), a new semi-empirical MP2-based method for dimer interaction-energy calculations. To correct for errors in MP2, SNS-MP2 uses quantum chemical features of the complex under study in conjunction with a neural network to reweight terms appearing in the total MP2 interaction energy. The method has been trained on a new data set consisting of over 200 000 complete basis set (CBS)-extrapolated coupled-cluster interaction energies, which are considered the gold standard for chemical accuracy. SNS-MP2 predicts gold-standard binding energies of unseen test compounds with a mean absolute error of 0.04 kcal mol−1 (root-mean-square error 0.09 kcal mol−1), a 6- to 7-fold improvement over MP2. To the best of our knowledge, its accuracy exceeds that of all extant density functional theory- and wavefunction-based methods of similar computational cost, and is very close to the intrinsic accuracy of our benchmark coupled-cluster methodology itself. Furthermore, SNS-MP2 provides reliable per-conformation confidence intervals on the predicted interaction energies, a feature not available from any alternative method.Keywords
This publication has 63 references indexed in Scilit:
- Psi4: an open‐source ab initio electronic structure programWIREs Computational Molecular Science, 2011
- S66: A Well-balanced Database of Benchmark Interaction Energies Relevant to Biomolecular StructuresJournal of Chemical Theory and Computation, 2011
- Parameterization of a B3LYP Specific Correction for Noncovalent Interactions and Basis Set Superposition Error on a Gigantic Data Set of CCSD(T) Quality Noncovalent Interaction EnergiesJournal of Chemical Theory and Computation, 2011
- Dispersion-corrected Møller–Plesset second-order perturbation theoryThe Journal of Chemical Physics, 2009
- On the accuracy of spin-component-scaled perturbation theory (SCS-MP2) for the potential energy surface of the ethylene dimerMolecular Physics, 2009
- Explicitly correlated RMP2 for high-spin open-shell reference statesThe Journal of Chemical Physics, 2008
- Truncation of the correlation consistent basis sets: Extension to third-row (Ga–Kr) moleculesThe Journal of Chemical Physics, 2005
- Truncation of the correlation consistent basis sets: An effective approach to the reduction of computational cost?The Journal of Chemical Physics, 2004
- Electron affinities of the first-row atoms revisited. Systematic basis sets and wave functionsThe Journal of Chemical Physics, 1992
- On the connection between the supermolecular Møller-Plesset treatment of the interaction energy and the perturbation theory of intermolecular forcesMolecular Physics, 1988