[gmx-users] Re: How to control the density

[Justin A. Lemkul]

teklebrh at ualberta.ca wrote:
> Dear Justin,
> 
>>  To use pdb2gmx, you need an .rtp entry.  For something like
>>  united-atom n-heptane, this should be a relatively easy exercise,
>>  even to write the topology by hand.  See the manual.
> 
> I just check the ffG53a6.rtp and I found that DPPC have an alkane chain. 
> In my case          n-heptane. I followed the same idea you gave me for 
> generating the toluene topology from PHE.  I made an entry to the .rtp 
> using DPPC.
> 
>   pdb2gmx -f Heptane.pdb -o Heptane.gro -p Heptane.top -missing
> 
>   I got the topology from this force filed. Is this RIGHT!
> 

The Gromos96 parameter sets do not reproduce dihedral properties of n-alkanes 
particularly well.  You'll have to go into the literature to see if this is an 
adequate model.  As far as tricking pdb2gmx, there's nothing wrong with what 
you've done, but you do need to justify that the resulting parameters are 
reasonable.

> Second, I am using the GROMOS96 ... ffG43a1 for my entire MD simulation 
> since I am using as well PRODRG topology file for my polymors. In order 
> to be consistent in all my simulation. I am using the ffG43a1 force 
> field for heptane. Now I took the charge and charge group ONLY from 
> ffG53a6.rtp of the DPPC and encorporate into the PRODRG.. to get a 
> better topology. Is this the right thing to do. Please comment
> 

For a united-atom n-alkane, it should have little effect.  All the carbons have 
zero charge.  But again I caution you about the model from which you are making 
the assumptions.  There is a recent parameterization of various lipids from 
Kukol, published some time ago in JCTC.  These may be better parameters, but I 
have not looked thoroughly through them.

> Third, how to avoid using a large tau_p. Like tau_p=3 for example. I 
> tried tau_p=1 but it does not look working. I got this error.
> 

Better equilibration prior to applying pressure coupling, probably.

> Fatal error:
> Too many LINCS warnings (1663)
> If you know what you are doing you can adjust the lincs warning 
> threshold in your mdp file
> or set the environment variable GMX_MAXCONSTRWARN to -1,
> but normally it is better to fix the problem
> 
> But as soon as i changed its value. The MD is still running. Please comment.
> 

I already did:

http://lists.gromacs.org/pipermail/gmx-users/2010-March/049411.html

> Fouth, about the solvent box. when I fill my box that contain 8 
> polymers. What density do i need to keep for my mixture. By changing the 
> solvent box I can change the entire density of the BOX of the actual 
> simulation. what is the correct way of handling density.
> 

The density depends largely on the contents of the system.  If you have a high 
percentage of solute, then you approach some sort of binary mixture rather than 
a solution.  At very low solute concentration, the density of the solvent will 
dominate, and so that is the "right" density, more or less.  The topic of 
density has come up several times already this week; the ability to reproduce 
experimental density depends on the quality of the model you employ.  You may 
not achieve a perfectly correct experimental density without going through the 
painful process of parameterization:

http://www.gromacs.org/Documentation/How-tos/Parameterization

If you rely on stock parameters to do the job for you, you are subject to the 
model's inherent limitations and flaws.

-Justin

> Regards,
> 
> R
> 
> 
> Quoting "Justin A. Lemkul" <jalemkul at vt.edu>:
> 
>  >
>  >
>  > teklebrh at ualberta.ca wrote:
>  >> Dear Justin.
>  >>
>  >> I turned off the gen-vel to no. Yes you are right. But I got the 
> same error.
>  >>
>  >> I increase tau-p ( from 1 to 3) and it looks working fine so far.
>  >>
>  >> One of the warnings is that the pressure is less 1%.
>  >>
>  >
>  > It could be that your previous equilibration was a bit insufficient
>  > to relax whatever the system is so there are still some colliding
>  > particles that cause the temperature to spike.  Increasing tau_p
>  > makes the coupling a bit weaker, allowing for some greater
>  > fluctuation, but you still may run into problems.
>  >
>  >> Thank you for your help Justine.
>  >>
>  >> ================================
>  >>
>  >> Remark from the previous request
>  >>
>  >> The pdb2gmx is also working .... the reinstalled the .rtp file and 
> it is ok!
>  >>
>  >> I got the Toluene topology from the PHE.
>  >>
>  >> ===================================
>  >>
>  >> I want the same for the n-heptane toplogy using the pdb2gmx.... is
>  >> there any method I could get that from the gromos96 force filed.
>  >> Just to avoid using PRODRG!
>  >>
>  >
>  > To use pdb2gmx, you need an .rtp entry.  For something like
>  > united-atom n-heptane, this should be a relatively easy exercise,
>  > even to write the topology by hand.  See the manual.
>  >
>  > -Justin
>  >
>  > --
>  > ========================================
>  >
>  > Justin A. Lemkul
>  > MILES-IGERT Trainee
>  > Department of Biochemistry
>  > Virginia Tech
>  > Blacksburg, VA
>  > jalemkul[at]vt.edu | (540) 231-9080
>  > http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
>  >
>  > ========================================
>  > --
>  > gmx-users mailing list    gmx-users at gromacs.org
>  > http://lists.gromacs.org/mailman/listinfo/gmx-users
>  > Please search the archive at http://www.gromacs.org/search before 
> posting!
>  > Please don't post (un)subscribe requests to the list. Use the www
>  > interface or send it to gmx-users-request at gromacs.org.
>  > Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>  >
>  >
> 

-- 
========================================

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================