[gmx-users] Re: CHARMM TIP3-Water with GMX

[Pär Bjelkmar]

Hi Stephane,

> I am doing some tests with the CHARMM port in GROMACS. Before to start 
> more extensive simulations with this ff
> I have performed a short run of 1000 TIP3-CHARMM water molecules in a 
> cubic during at T=300 K and P=1.015 bar with v-rescale and PR as 
> thermostat and barostat. I have used the VERSION 
> 4.0.99_development_20090927 of GMX.

good to hear that you could reproduce our results.

> 
> Moreover, since I am not a power user of  GROMACS, I don't know how to 
> translate in GROMACS directives the sentence given in JCTC paper " using 
> a short-range cutoff of 1.2 nm, and van der Waals interactions were 
> switched off between 1.0 to 1.2 nm".
This means coulombtype = PME, rcoulomb = 1.2, vdw-type = switch, rvdw-switch = 1.0, rvdw = 1.2. 

> And by consequence what are the 
> corrects coulomb and vdw parameters for simulations of CHARMM ff ?
This you should be able to figure out from the CHARMM ff papers. According to the CMAP paper (MacKerell et al. Journal of Computational Chemistry 2004) they used real-space (short-range) cutoff of their electrostatic calculations (using PME) of 8.5 Å, and the LJ were switched of between 7.5 to 8.5 Å and their neighbor lists had a cutoff of 10.5 Å. Remember that we performed the testing (comparing the energy terms) in the paper without cut-offs.

/Pär