Re: Problems with wrapAll

From: Vermaas, Joshua (Joshua.Vermaas_at_nrel.gov)
Date: Thu Jul 21 2016 - 11:14:48 CDT

Hi Sonu,

It looks like your cell basis vectors don’t match what you describe. Since they are a little smaller, if you get unlucky and a bond is wrapped across the periodic boundary, wrapall will dutifully wrap everything so that the relative starting configuration is retained, which includes a really long bond. What I do to avoid this is overestimate the cell basis vectors somewhat, and let constant pressure MD shrink the box size to the appropriate size.

Your cutoffs will need to be changed for production simulation. Charmm was parameterized against specific cutoffs, and while the simulations will be ALOT faster with shorter cutoffs, the physical properties of the system are not going to be retained with much shorter cutoffs. You may need to make your simulation box larger, as the polymer may start to interact with itself when it rotates.

Finally, we aren’t all sirs on this mailing list, and sometimes that bears repeating.

-Josh

On Jul 20, 2016, at 11:52 PM, Sonu Kumar <sonu2kd_at_gmail.com<mailto:sonu2kd_at_gmail.com>> wrote:

Sir,
I have been trying to do a md simulation for my polymer with the help of NAMD in a periodic box of 24.955 Angstroms.
For this,
I took the minimized molecule and wrapped it with the help of VMD along the center of mass in a box of 24.955 A. I extracted the pdb for this and used this file in simulation so that no bond crosses the boundary of the periodic box.

Then I ran the md simulation for this pdb file with configuration listed at the end. However, the output which I received showed bonds sticking out despite the fact WRAPALL WAS ON(just emphasizing). I tried everything like increasing or decreasing minimization or doing it in many steps wrapping it along every step.
However, everytime bonds sticked out of the box.

Can you help me in solving this problem of why wrapAll is not working. Moreover, is it alright to choose lesser values of cutoff, switchdist and pairlistdist as it was showing patch grid error for 25 A box.

Configuration file :

structure final.psf
coordinates haha1.pdb
outputName sonu12345

set temperature 298

firsttimestep 0

paraTypeCharmm on
parameters par_all22_prot.prm

temperature $temperature

cellBasisVector1 17.69565 0 0
cellBasisVector2 0 18.89368 0
cellBasisVector3 0 0 24.77155
cellOrigin 5.054398 6.476589 12.377248

#wrapWater on
wrapAll on

exclude scaled1-4
1-4scaling 1.0
cutoff 4.9
switching on
switchdist 4.0
pairlistdist 5.0
#Chose lesser values because of patch grid error for 25 A box

# Integrator Parameters
timestep 2.0 ;# 2fs/step
rigidBonds none ;# needed for 2fs steps
nonbondedFreq 1
fullElectFrequency 2
stepspercycle 10

langevin on ;# do langevin dynamics
langevinDamping 1 ;# damping coefficient (gamma) of 5/ps
langevinTemp $temperature
langevinHydrogen no ;# don't couple langevin bath to hydrogens

restartfreq 500 ;# 500steps = every 1ps
dcdfreq 500
xstFreq 500
outputEnergies 100
outputPressure 100

minimize 1000
reinitvels $temperature

run 100000 ;# 200ps

This archive was generated by hypermail 2.1.6 : Tue Dec 27 2016 - 23:22:19 CST