C.R. Sweet, S.S. Hampton,
R.D. Skeel, J.a. Izaguirre


SepArable shadow Hamiltonian hybrid Monte Carlo method

supplementary information

 

Introduction. 2

Additional results

Simulation and implementation details. 3

Input files. 3

Analysis scripts. 3

Executable

 


Introduction


This supplementary information contains the input files, executable and analysis scripts to reproduce the results presented in the ‘Separable Shadow Hamiltonian Hybrid Monte Carlo method’ paper. 

 


Additional results


Faster convergence to the target average is expected from S2HMC when compared to HMC. This is illustrated in the following figure for a box of 4002 water molecules.

convergepe2.jpg

 


Simulation and implementation details


Input files

Input files for box of 1002 water molecules.
Charmm par file
Charmm psf file
XYZ position file
Protomol HMC conf file
Protomol S2HMC conf file

Input files for box of 2001 water molecules.
Charmm par file
Charmm psf file
XYZ position file
Protomol HMC conf file
Protomol S2HMC conf file

Input files for box of 4002 water molecules.
Charmm par file
Charmm psf file
XYZ position file
Protomol HMC conf file
Protomol S2HMC conf file

Input files for box of 8002 water molecules.
Charmm par file
Charmm psf file
XYZ position file
Protomol HMC conf file
Protomol S2HMC conf file

Input files for box of 16002 water molecules.
Charmm par file
Charmm psf file
XYZ position file
Protomol HMC conf file
Protomol S2HMC conf file

 

 

Analysis scripts

Calculate average/estimated variance and weights
Takes the Protomol energies file and calculates the mean and estimated variance of the selected parameter and the weights: [av1,var1,w]=calcWeightsAv('w4002s2hmc',300,2) loads file ‘w4002s2hmc’ sets the temperature to 300’K and finds the average and estimated variance for parameter 2 (potential energy).

 

Protomol executable.

protomol Protomol executable for Linux/Intel.