[gmx-users] Re: charmm lipid simulation and vsites

[Pär Bjelkmar]

Hello David,

I send answer to the gromacs mailing list also.

12 apr 2010 kl. 17.51 skrev David Caplan:
> I am still experimenting with the CHARMM27 support in Gromacs. I am
> using a very recent git version (from last week).
> 
> Currently, I'm trying to equilibrate a membrane bilayer with virtual
> sites. The membrane is DPPC with the Sonne et al. parameters (for zero
> surface tension):
> Sonne et al. Reparameterization of all-atom
> dipalmitoylphosphatidylcholine lipid parameters enables simulation of
> fluid bilayers at zero tension. Biophys J (2007) vol. 92 (12) pp.
> 4157-67
> 
> I am not so familiar with vsites, but I have read the documentation
> and I saw you used vsites in your recent paper. Have you used vsites
> with membrane simulations using CHARMM27?
No, I have not but it should be straight forward.

> If so, what kind of pressure
> coupling do you use? is any surface tension applied?
> If you have an example MDP that you used for this I would love to take
> a look at it.
I would use semiisotropic Berendsen for equilibration and perhaps switch to Parrinello-Rahman during production but when it comes to surface tension and CHARMM simulations I'm afraid I don't have any experience. Guess you have to read the cited paper carefully to figure this out. 

> 
> Also, do you typically run your CHARMM27 simulations using Langevin dynamics?
No, not when you have explicit solvent. 

/Pär