Gaussian Process-Based Refinement of Dispersion Corrections.

08:00 EDT 11th October 2019 | BioPortfolio

Summary of "Gaussian Process-Based Refinement of Dispersion Corrections."

We employ Gaussian process (GP) regression to adjust for systematic errors in D3-type dispersion corrections. We refer to the associated, statistically improved model as D3-GP. It is trained on differences between interaction energies obtained from PBE-D3(BJ)/ma-def2-QZVPP and DLPNO-CCSD(T)/CBS calculations. We generated a data set containing interaction energies for 1248 molecular dimers, which resemble the dispersion-dominated systems contained in the S66 data set. Our systems represent not only equilibrium structures but also dimers with various relative orientations and conformations at both shorter and longer distances. A reparametrization of the D3(BJ) model based on 66 of these dimers suggests that two of its three empirical parameters, and , are zero, whereas = 5.6841 bohr. For the remaining 1182 dimers, we find that this new set of parameters is superior to all previously published D3(BJ) parameter sets. To train our D3-GP model, we engineered two different vectorial representations of (supra-)molecular systems, both derived from the matrix of atom-pairwise D3(BJ) interaction terms: (a) a distance-resolved interaction energy histogram, histD3(BJ), and (b) eigenvalues of the interaction matrix ordered according to their decreasing absolute value, eigD3(BJ). Hence, the GP learns a mapping from D3(BJ) information only, which renders D3-GP-type dispersion corrections comparable to those obtained with the original D3 approach. They improve systematically if the underlying training set is selected carefully. Here, we harness the prediction variance obtained from GP regression to select optimal training sets in an automated fashion. The larger the variance, the more information the corresponding data point may add to the training set. For a given set of molecular systems, variance-based sampling can approximately determine the smallest subset being subjected to reference calculations such that all dispersion corrections for the remaining systems fall below a predefined accuracy threshold. To render the entire D3-GP workflow as efficient as possible, we present an improvement over our variance-based, sequential active-learning scheme [ 2018 , 14 , 5238 ]. Our refined learning algorithm selects multiple (instead of single) systems that can be subjected to reference calculations simultaneously. We refer to the underlying selection strategy as batchwise variance-based sampling (BVS). BVS-guided active learning is an essential component of our D3-GP workflow, which is implemented in a black-box fashion. Once provided with reference data for new molecular systems, the underlying GP model automatically learns to adapt to these and similar systems. This approach leads overall to a self-improving model (D3-GP) that predicts system-focused and GP-refined D3-type dispersion corrections for any given system of reference data.


Journal Details

This article was published in the following journal.

Name: Journal of chemical theory and computation
ISSN: 1549-9626


DeepDyve research library

PubMed Articles [20187 Associated PubMed Articles listed on BioPortfolio]

Gaussian Process Trajectory Learning and Synthesis of Individualized Gait Motions.

This paper proposes a Gaussian process-based method for trajectory learning and generation of individualized gait motions at arbitrary user-designated walking speeds, intended to be used in generating...

Small Basis Set Allowing the Recovery of Dispersion Interactions with Double-Hybrid Functionals.

Taking advantage of the compensation between Basis Set Superposition Error and Basis Set Incompleteness Error, a new basis is developed to improve the performances of Double Hybrid (DH) functionals in...

One-way coupling of WRF with a Gaussian dispersion model: a focused fine-scale air pollution assessment on southern Mediterranean.

Numerous uncertainty factors in dispersion models should be taken into account in order to improve the reliability of predictions. The ability of a mesoscale meteorological model to assimilate observa...

Immediate-released pelletized solid dispersion containing fenofibrate: formulation, in vitro characterization, and bioequivalence studies in experimental beagle dogs.

There have been many strategies to increase solubility, dissolution rates, and oral bioavailability of fenofibrate such as micronization, nanonization, solid dispersion, and emulsion so far. To our kn...

Information fusion estimation-based path following control of quadrotor UAVs subjected to Gaussian random disturbance.

Random disturbance has a detrimental effect on the reliability and safety of quadrotor unmanned aerial vehicles (UAVs). This paper proposes an anti-Gaussian random disturbance control method for the p...

Clinical Trials [4961 Associated Clinical Trials listed on BioPortfolio]

A Randomized Controlled Trial of HIV Testing and Linkage to Care at Community Corrections

The investigators propose to conduct both a randomized trial of HIV testing in community corrections, and a randomized trial of linkage to HIV care for people with HIV recruited through co...

Pigment Dispersion Syndrome: Natural History and Possible Protective Effect of a YAG Laser Iridotomy

STUDY AIMS 1. To determine the 10-year conversion rate from pigment dispersion syndrome (PDS) to pigmentary glaucoma (PG) 2. To evaluate the possible protective effect of ...

The Effects of Different Anesthetic Techniques on QT, Corrected QT (QTc), and P Wave Dispersions in Cesarean Section

This study evaluates the effects of different anesthetic techniques on QT, QTc, and Pwd in cesarean section. Half of participants received general anesthesia, while the other half received...

Innovative Liver Elasticity, Attenuation, and Dispersion Ultrasound Study

The objective of this study is: (1) to investigate the correlation of ultrasound parameters (SW speed, Dispersion slope, Attenuation value, Normalized Local Variance, Liver / Kidney Intens...

Relative Bioavailability Study of IX-01 Caplet Versus Aqueous Dispersion and Food Effect of the Caplet in Healthy Males

An open-label, randomized, three-period, three-way crossover trial of single doses of IX-01 in 12 healthy male subjects. In each period, subjects will receive a single oral dose of 1600 mg...

Medical and Biotech [MESH] Definitions

The method of measuring the dispersion of an optically active molecule to determine the relative magnitude of right- or left-handed components and sometimes structural features of the molecule.

Process of classifying cells of the immune system based on structural and functional differences. The process is commonly used to analyze and sort T-lymphocytes into subsets based on CD antigens by the technique of flow cytometry.

A non-medical term defined by the lay public as a food that has little or no preservatives, which has not undergone major processing, enrichment or refinement and which may be grown without pesticides. Health foods have been attributed with the ability to prevent the development of diseases, slow the aging process, and prolong life. (from Segen, The Dictionary of Modern Medicine, 1992)

A direct form of psychotherapy based on the interpretation of situations (cognitive structure of experiences) that determine how an individual feels and behaves. It is based on the premise that cognition, the process of acquiring knowledge and forming beliefs, is a primary determinant of mood and behavior. The therapy uses behavioral and verbal techniques to identify and correct negative thinking that is at the root of the aberrant behavior.

Process of determining and distinguishing species of bacteria or viruses based on antigens they share.

Quick Search

DeepDyve research library

Searches Linking to this Article