Facilitating <i>ab initio</i> QM/MM free energy simulations by Gaussian process regression with derivative observations
Ryan Snyder, Bryant Kim, Xiaoliang Pan, Yihan Shao, Jingzhi Pu
Abstract
In this machine-learning-facilitated method, Gaussian process regression (GPR) is used to predict energy and force corrections for a semiempirical QM/MM level to match with ab initio QM/MM results during MD-based free energy simulations.
Topics & Concepts
KrigingAb initioGaussian processGaussianQM/MMEnergy (signal processing)RegressionComputational chemistryAb initio quantum chemistry methodsDerivative (finance)Process (computing)Statistical physicsChemistryPhysicsMolecular dynamicsComputer scienceMathematicsQuantum mechanicsMachine learningMoleculeStatisticsOperating systemFinancial economicsEconomicsMachine Learning in Materials ScienceProtein Structure and DynamicsComputational Drug Discovery Methods