Bill - Thanks for your response. The way I created the system is a little
bit complicated but let me explain it as much as can.
I used leap to first solvate the system. There are 2 planes parallel to
each other in the system (both parallel to xy plane) located at z1 and z2.
There is no water above z2 or below z1. This is a periodic system; so for
instance a crystal can be formed if these parallel layers are repetitevly
put on top of each other.
Then, I randomly distributed 139 Na+ and Cl- ions by removing water
molecules. I used the Cheatham's revised Na+ and Cl- parameters.
Initially, I tried to use AddToBox for this purpose but this created much
much bigger vacuums in the system. There are almost 5811 water molecules
in the system.
I do 2 stage equilibration; first NVT then NPT.
------ mdeq1 --------
Equilibration of the whole system: Temperature (10 ps)
&cntrl
imin = 0, irest = 0, ntx = 1,
ntb = 1, cut = 8.0, ntr = 1,
ntc = 2, ntf = 2,
tempi = 0.0, temp0 = 300.0, ntt = 3, gamma_ln = 1.0, ig=6,
nstlim = 5000, dt = 0.002,
ntpr = 100, ntwx = 1000, ntwr = 1000
/
Hold the structure fixed
500.0
RES 1 3022
END
END
--------------------
This will increase the temperature to 300K in 5000 steps. Then I do the
NPT runs:
----- mdeq2 ---------
Equilibration of the whole system: Pressure
&cntrl
imin = 0, irest = 1, ntx = 5,
ntb = 2, pres0 = 1.0, ntp = 1, taup = 2.0,
cut = 8.0, ntr = 0,
ntc = 2, ntf = 2,
tempi = 300.0, temp0 = 300.0, ntt = 3, gamma_ln = 1.0, ig=6,
nstlim = 1000000, dt = 0.001,
ntpr = 100, ntwx = 1000, ntwr = 1000
/
----------------------
This still creates bubbles in the system. Is it possible that the number
of water molecules is not enough to bring the system to the expected
conditions? The bubbles are formed at the sides of the system (not around
a particular ion). The system's Na+ concentration is almost 1 M.
In order to visualize, what I do is to use ptraj to bring the atoms into
unit cell, which shows these vacuum bubbles.
--------- ptraj.in ----------
trajin mdeq4.rst
trajout pdb out pdb
center :1-3022
image familiar
go
----------------------------
Thanks again. Best regards,
PS: If it might help you see how the system looks like, I can send you the
structure in private.
Ilyas Yildirim, Ph.D.
-----------------------------------------------------------
= Department of Chemistry - 2145 Sheridan Road =
= Northwestern University - Evanston, IL 60208 =
= Ryan Hall #4035 (Nano Building) - Ph.: (847)467-4986 =
=
http://www.pas.rochester.edu/~yildirim/ =
-----------------------------------------------------------
On Wed, 3 Nov 2010, Bill Ross wrote:
>> Yes, the threads in the mailing list always says that NPT has to be used
>> in order to get rid of vacuum bubbles. I do use NPT in the equilibration
>> process. The problem is dominant when I have a salt concentration around 1
>> M. In a test system, when I do not have any salt at all, the system looks
>> pretty reasonable after the equilibration. When the system has 1 M of salt
>> in it, however, water molecules are sucked such that it creates vacuum
>> bubbles. And they get bigger even after continuing the simulation with
>> NPT.
>
> It might be useful to see all your equilibrateion mdin files. I'd
> say the NPT should be run until density reaches 1, then switch to
> NVT and the bubbles can't form :-) It might be convenient to do
> this at lower temperature, though this problem might be worth
> understanding better before moving on.
>
> What salt are you using? Do the vacuum bubbles form at the ions?
> How many waters, how many ions?
>
>> I was going to send a second email regarding vacuum bubble problem I am
>> having in my system, but one of my collegue said that it might be a
>> problem with the box information. As a result, I wanted to understand how
>> sander and pmemd use the boxinfo in periodicity and calculations.
>
> Did you solvate with leap? No matter what the box is, NPT should fix
> any vacuum.
>
> Bill
>
> _______________________________________________
> 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 Nov 03 2010 - 18:00:03 PDT
