Tutorial of using GaMD in Amber

Alanine Dipeptide:

GaMD Theory

A PDF tutorial of running GaMD simulations on alanine dipeptide: GaMD_Amber-Tutorial.pdf

Source files of the GaMD-Amber tutorial: amber-gamd-tutorial.tgz (30.2MB)

 Note: While the NVT ensemble has been used in the tutorial simulations with "ntb=1", the NPT ensemble is generally preferred for running GaMD simulations (especially for the GaMD equilibration) with parameters like "ntb = 2, ntp = 1, barostat = 1".

Example input parameters for running GaMD in Amber:

In case you are running GaMD simulation on a computer cluster that imposes wallclock limit for simulation jobs, you may divide your long simulation into multiple jobs, which could include for example job #1 for 2 ns conventional MD (cMD) and 50 ns GaMD equilibration, job #2 for starting GaMD production simulation with radomized atomic velocities, and job #3 for extending GaMD production simulation until the end:

Job #1: cMD and GaMD equilibration

nstlim = 26000000,
irest = 0,
ntx = 1,


igamd = 3, iE = 1, irest_gamd = 0,
ntcmd = 1000000, nteb = 25000000, ntave = 200000,
ntcmdprep = 200000, ntebprep = 800000,
sigma0P = 6.0, sigma0D = 6.0,



Job #2: start GaMD production simulation

nstlim = 50000000,
irest = 0,
ntx = 1,


igamd = 3, iE = 1, irest_gamd = 1,
ntcmd = 0, nteb = 0, ntave = 200000,
ntcmdprep = 0, ntebprep = 0,
sigma0P = 6.0, sigma0D = 6.0,

Job #3: repeat running jobs using this input file until end of your GaMD production simulation

nstlim = 50000000,
irest = 1,
ntx = 5,


igamd = 3, iE = 1, irest_gamd = 1,
ntcmd = 0, nteb = 0, ntave = 200000,
ntcmdprep = 0, ntebprep = 0,
sigma0P = 6.0, sigma0D = 6.0,

Note: You can also perform multiple independent GaMD simulations with randomized atomic velocities by repeating jobs #2 and #3.

 

LiGaMD in Amber

Example input parameters used in dual-boost LiGaMD simulations include the following in addition to those used in conventional MD:

igamd = 11, irest_gamd = 0,
ntcmd = 700000, nteb = 27300000, ntave = 140000,
ntcmdprep = 280000, ntebprep = 280000,
sigma0P = 4.0, sigma0D = 6.0, iEP = 2, iED=1,

icfe = 1, ifsc = 1, gti_cpu_output = 0, gti_add_sc = 1,
timask1 = ':225', scmask1 = ':225',
timask2 = '', scmask2 = '',

ibblig = 1, nlig = 10, atom_p = 2472, atom_l = 4, dblig = 3.7,


OR

igamd = 11, irest_gamd = 0,
ntcmd = 700000, nteb = 27300000, ntave = 140000,
ntcmdprep = 280000, ntebprep = 280000,
sigma0P = 4.0, sigma0D = 6.0, iEP = 2, iED=1,

icfe = 1, ifsc = 1, gti_cpu_output = 0, gti_add_sc = 1,
timask1 = ':225', scmask1 = ':225',
timask2 = '', scmask2 = '',

ibblig = 2, nlig = 10, atom_l = 4,
ntmsd = 50000, nftau = 10, dblig = 20,


Pep-GaMD in Amber

Example input parameters used in Pep-GaMD simulations include the following in addition to those used in conventional MD:

igamd = 15, iE = 1, iEP = 1, iED = 1, irest_gamd = 0,
ntcmd = 1000000, nteb = 1000000, ntave = 50000,
ntcmdprep = 200000, ntebprep = 200000,
sigma0P = 6.0, sigma0D = 6.0,

icfe = 1, ifsc = 1, gti_cpu_output = 0, gti_add_sc = 1,
timask1 = ':225', scmask1 = ':225',
timask2 = '', scmask2 = '',



Copyright Reserved © Miao Lab, University of Kansas
◇ Last updated: Mon, July 18, 2022