2017
DOI: 10.1063/1.4983164
|View full text |Cite
|
Sign up to set email alerts
|

Estimating thermodynamic expectations and free energies in expanded ensemble simulations: Systematic variance reduction through conditioning

Abstract: Markov chain Monte Carlo methods are primarily used for sampling from a given probability distribution and estimating multi-dimensional integrals based on the information contained in the generated samples. Whenever it is possible, more accurate estimates are obtained by combining Monte Carlo integration and integration by numerical quadrature along particular coordinates. We show that this variance reduction technique, referred to as conditioning in probability theory, can be advantageously implemented in exp… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
12
0

Year Published

2019
2019
2022
2022

Publication Types

Select...
5
3

Relationship

0
8

Authors

Journals

citations
Cited by 10 publications
(12 citation statements)
references
References 63 publications
(122 reference statements)
0
12
0
Order By: Relevance
“…Moreover, free energy calculations commonly involve simulating the same system in multiple intermediate states-which are not always physical intermediates-that do not necessarily have the same kinetic properties. Commonly, quantitative comparisons of performance are based on the standard deviation of the free energy estimates after roughly the same computational cost [37][38][39][40]. This statistic, however, does not quantify the bias, which is, in general, not negligible.…”
Section: Comparing the Efficiency Of Methods Requires Eliminating Conmentioning
confidence: 99%
“…Moreover, free energy calculations commonly involve simulating the same system in multiple intermediate states-which are not always physical intermediates-that do not necessarily have the same kinetic properties. Commonly, quantitative comparisons of performance are based on the standard deviation of the free energy estimates after roughly the same computational cost [37][38][39][40]. This statistic, however, does not quantify the bias, which is, in general, not negligible.…”
Section: Comparing the Efficiency Of Methods Requires Eliminating Conmentioning
confidence: 99%
“…Moreover, free energy calculations commonly involve simulating the same system in multiple intermediate states-which are not always physical intermediates-that do not necessarily have the same kinetic properties. Commonly, quantitative comparisons of performance are based on the standard deviation of the free energy estimates after roughly the same computational cost [34][35][36][37]. This approach, however, has a few deficiencies.…”
Section: We Need Robust General Strategies To Measure the Efficiency mentioning
confidence: 99%
“…Specific variants of this adaptive approach include self-adjusted mixture sampling (SAMS) , and the accelerated weight histogram method . Most other free energy approaches rely on multiple parallel or coupled simulations performed for each alchemical intermediate, as in Hamiltonian replica exchange. , Therefore, the ability to sample multiple ensembles in a single simulation replica is attractive for many applications, such as enabling large-scale virtual screening on distributed parallel computing platforms. EE methods, however, have not been widely adopted for estimating ligand binding free energies per se, although much work has been done toward improving adaptive expanded ensemble estimates for calculating free energies from perturbed Hamiltonians. …”
Section: Introductionmentioning
confidence: 99%