Work

Hamiltonian Monte Carlo with Constrained Molecular Dynamics as Gibbs Sampling

Year: 2017

Type: article

Abstract: Compared to fully flexible molecular dynamics, simulations of constrained systems can use larger time steps and focus kinetic energy on soft degrees of freedom. Achieving ergodic sampling from the Boltzmann distribution, however, has proven challenging. Using recent generalizations of the equipartition principle and Fixman potential, here we implement Hamiltonian Monte Carlo based on constrained molecular dynamics as a Gibbs sampling move. By mixing Hamiltonian Monte Carlo based on fully flexible and torsional dynamics, we are able to reproduce free energy landscapes of simple model systems and enhance sampling of macrocycles.

Source: Journal of Chemical Theory and Computation

DOI: 10.1021/acs.jctc.7b00570

Authors: Laurentiu Spiridon, David D. L. Minh

Institutions: Illinois Institute of Technology, Institute of Biochemistry, Romanian Academy

Language: English

Topic: Markov Chains and Monte Carlo Methods

Subfield: Statistics and Probability

Field: Mathematics

Domain: Physical Sciences

Open Access status: green

Funders: National Institutes of Health, National Institute of General Medical Sciences

Award: Sound-stage Virtual Screening Based on Implicit Ligand Theory