Exact non-Markovian permeability from rare event simulations

Ghysels, An; Roet, Sander; Davoudi, Samaneh; Erp, Titus S. van

doi:10.1103/physrevresearch.3.033068

Cited by 16 publications

(49 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While from the average fraction it still looks like the simulations sampled both channels for any number of workers, 4 out of the 5 simulations in the K = N = 12 case solely visited one of the two channels. This is in agreement with previous TIS results [29]. The K = 11 case already provides a dramatic improvement, but is still expected to be sub optimal due to the relatively low frequency of replica exchange moves compared to K < 11.…”

Section: Two-channel Simulationssupporting

confidence: 92%

“…All simulations are somewhat higher, though still in good agreement with the previous simulation from Ref. [29]. We also evaluated the approximate result based on Kramers' theory (see SI) which seem to confirm the results obtained in this paper.…”

Section: Two-channel Simulationssupporting

confidence: 89%

“…While this artificial simulation allows us to investigate the potential strength of the method to tackle extremely rare events, it cannot reveal the effect of correlations between accepted paths when fast exploration of the reaction coordinate's orthogonal directions are crucial. To analyze this effect, we also ran a 2D membrane permeation system with two slightly asymmetric channels [29]. Lastly, to study our algorithm with a more generic W -matrix that needs to be solved via BBFG formula, we also ran a set of underdamped Langevin simulations of a particle in a double well potential [30] using the recent wire fencing algorithm with the high acceptance protocol [26].…”

Section: Resultsmentioning

confidence: 99%

“…As reported in ref. [29], this ratio requires many MC moves to converge to the theoretical value of 0.71 without the added MC moves introduced in that paper. We did not simulate with these added moves and thus see the same slow convergence for all of our simulations.…”

Section: Two-channel Simulationsmentioning

confidence: 97%

“…The wall-time efficiency shows an upward trend until 10 workers after which it starts to drop off as well. We assign this drop to the reduction of replica exchange moves which is an essential aspect for sampling this system efficiently [29]. This is tangible from figure S1 in the SI where we plot fraction of trajectories, passing through λ M−1 , that are in the lower barrier channel.…”

Section: Two-channel Simulationsmentioning

confidence: 99%

See 4 more Smart Citations

Exchanging replicas with unequal cost, infinitely and permanently

Roet¹,

Zhang²,

Erp³

2022

Preprint

Self Cite

View full text Add to dashboard Cite

We developed a replica exchange method that is effectively parallelizable even if the computational cost of the Monte Carlo moves in the parallel replicas are considerably different, for instance, because the replicas run on different type of processor units or because of the algorithmic complexity. To prove detailed-balance, we make a paradigm shift from the common conceptual viewpoint in which the set of parallel replicas represents a high-dimensional superstate, to an ensemble based criterion in which the other ensembles represent an environment that might or might not participate in the Monte Carlo move. In addition, based on a recent algorithm for computing permanents, we effectively increase the exchange rate to infinite without the steep factorial scaling as function of the number of replicas. We illustrate the effectiveness of the replica exchange methodology by combining it with a quantitative path sampling method, replica exchange transition interface sampling (RETIS), in which the costs for a Monte Carlo move can vary enormously as paths in a RETIS algorithm do not have the same length and the average path lengths tend to vary considerably for the different path ensembles that run in parallel. This combination, coined ∞RETIS, was tested on three model systems.

show abstract

Section: Two-channel Simulationssupporting

confidence: 92%

Section: Two-channel Simulationssupporting

confidence: 89%

Section: Resultsmentioning

confidence: 99%

Section: Two-channel Simulationsmentioning

confidence: 97%

Section: Two-channel Simulationsmentioning

confidence: 99%

See 3 more Smart Citations

Exchanging replicas with unequal cost, infinitely and permanently

Roet¹,

Zhang²,

Erp³

2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

PyRETIS 3: Conquering rare and slow events without boundaries

Vervust,

Zhang,

Ghysels

et al. 2024

J Comput Chem

Self Cite

View full text Add to dashboard Cite

We present and discuss the advancements made in PyRETIS 3, the third instalment of our Python library for an efficient and user‐friendly rare event simulation, focused to execute molecular simulations with replica exchange transition interface sampling (RETIS) and its variations. Apart from a general rewiring of the internal code towards a more modular structure, several recently developed sampling strategies have been implemented. These include recently developed Monte Carlo moves to increase path decorrelation and convergence rate, and new ensemble definitions to handle the challenges of long‐lived metastable states and transitions with unbounded reactant and product states. Additionally, the post‐analysis software PyVisa is now embedded in the main code, allowing fast use of machine‐learning algorithms for clustering and visualising collective variables in the simulation data.

show abstract