A Julia simulation package with functions for running adaptive umbrella sampling Monte Carlo simulations of crosslinked actin rings.
This package implements the simulation protocol outlined in the Supplemental Material of Ref. 1.
The package can be installed by starting the Julia REPL, typing ] to enter package mode, and running
add ActinRingsMC
to install from the General registry, or by running
add https://github.com/cumberworth/ActinRingsMC.jl
to install directly from the development repository.
In examples directory of the repository is a script for umbrella sampling.
It has both an initial run and a continuation run.
The commands can be run in the Julia REPL, or, when in the same directory as the script, it can be run with
julia run-umbrella-sampling.jl
To achieve good sampling, one should experiment with the number of steps per iteration and the number of iterations.
The simulations output two data file types per simulation, or in the case of an umbrella sampling run, per iteration.
The file type with an .ops extension contains order parameters saved at steps determined by the write interval and can be read in as a dataframe for analysis and plotting.
Currently the order parameters include the energy, lattice height, and ring radius.
The file type with a .vtf extension is able to be read by the molecular visualization program, VMD, such that the configurations of the simulations can be spatially visualized.
When viewing in VMD, it is recommended to use "Chain" for "Coloring Method" and "VDW" for "Drawing Method".
The simulations also output a file containing all the system and simulation parameters used in the run to a .parms file.
Umbrella sampling runs output additional file types.
The output of these file is based on every step, not only those determined by the write frequency.
The top row gives the lattice height (or bin, although binning has not been tested).
Each subsequent row contains data from an iteration.
.counts gives the number of times each lattice height was visited, .freqs gives a normalized version of this.
.biases includes the biases that were used to for that iteration (in contrast to those that would be calculated from its data).
A related python package, actinrings, includes code for analyzing and plotting the output from these simulations, including free energies and ring constriction forces; see its documentation for details.
[1] A. Cumberworth and P. R. ten Wolde, Constriction of actin rings by passive crosslinkers, arXiv:2203.04260 [physics.bio-ph].