Re: [AMBER] Free Energy Calculations of Potassium and Chloride

From: Sai Kumar Ramadugu <sramadugu.gmail.com>
Date: Wed, 18 Jan 2012 12:43:25 -0600

Dr Steinbrecher,
I checked the input files and I am using the soft-core parameter, ifsc = 1.

Here is my input file for
V0:
NPT production
 &cntrl
  imin = 0, ntx = 5, irest = 1,
  ntpr = 10000, ntwr = 100000, ntwx = 10000,
  ntf = 1, ntc = 1,
  ntb = 2, cut = 10.0,
  nstlim = 500000, dt = 0.001,
  temp0 = 300.0, ntt = 3, gamma_ln = 2,
  ntr = 1, restraint_wt = 25.0,
  restraintmask = ':1-2'
  ntp = 1, pres0 = 1.0, taup = 2.0,
  icfe=1, clambda = 0.99,
  ifsc=1,
  crgmask=':1-2',
  scmask=':1-2',
 &end

Residues 1 and 2 happen to be K+ and Cl-.

For V1:

NPT production
 &cntrl
  imin = 0, ntx = 5, irest = 1,
  ntpr = 10000, ntwr = 100000, ntwx = 10000,
  ntf = 1, ntc = 1,
  ntb = 2, cut = 10.0,
  nstlim = 500000, dt = 0.001,
  temp0 = 300.0, ntt = 3, gamma_ln = 2,
  ntp = 1, pres0 = 1.0, taup = 2.0,
  icfe=1, clambda = 0.99,
  ifsc=1,
  crgmask='',
  scmask='',
 &end


Regarsd
Sai

On Wed, Jan 18, 2012 at 12:15 PM, <steinbrt.rci.rutgers.edu> wrote:

> Hi,
>
> > The Coulomb part is smooth, but when I come to the vander Waals part at
> > lambda = 0.96. 0.97, 0.98, 0.99, the deviations of dV/dL are really very
> > large and the associated rms in dV/dL is large too.
> > Below are the values pasted from my step2 (removal of VdW) for KCl in
> > water.
> >
> > lambda dV/dL rms
>
> > 0.96 -45.3181 25.7754
> > 0.97 -60.0310 34.3345
> > 0.98 -89.1323 51.1157
> > 0.99 -177.6619 106.7890
>
> Are you sure you are using softcore potentials on all disappearing atoms?
> This looks like the origin singularity problem that Sc should fix. Check
> your input files to see if ifsc etc. are really all set ok.
>
> Kind Regards,
>
> Thomas
>
> Dr. Thomas Steinbrecher
> formerly at the
> BioMaps Institute
> Rutgers University
> 610 Taylor Rd.
> Piscataway, NJ 08854
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Jan 18 2012 - 11:00:04 PST
Custom Search