FEASST: Free energy and advanced sampling simulation toolkit

Abstract

The Free Energy and Advanced Sampling Simulation Toolkit (FEASST) is a free, open-source, modular program to conduct molecular and particle-based simulations with Metropolis, Wang-Landau and Transition-Matrix Monte Carlo methods [1-7]. FEASST is implemented in C++ and may be imported as a module within Python 2 or 3.1 This document describes the initial public release version 1.0 with the following features: 1. Simulation techniques • Wang-Landau Monte Carlo • Transition-matrix Monte Carlo • Metropolis Monte Carlo 2. Thermodynamic ensembles • Grand canonical ensemble • Isothermal isobaric ensemble • Canonical ensemble 3. Advanced Monte Carlo algorithms • Parallel confguration swaps • Floppy box 4. Intermolecular interactions • Charged interactions with the Ewald summation • Lennard-Jones with different exponential parameters, long range corrections, Yukawa, force shifted and/or Gaussians • Hard spheres, soft spheres and square wells 5. Modern software • Interface with C++ or as a Python module • Open Multi-Processing (OpenMP) parallelization • Check points to save and restart simulations • Robust unit testing Many more features are planned for release in future versions of FEASST including expanded ensembles in temperature and alchemical transformations, confgurational bias, geometric cluster algorithm, aggregation volume bias, confnement, patchy particle potentials and Mayer-sampling Monte Carlo.