[gmx-users] Questions about parameters

[Warren Gallin]

   I have been working through some sample problems and now I am at the stage where I need to settle on a consistent setup for simulation of short peptides in aqueous solution.  I've been through the documentation repeatedly, but there are a couple things on which I am unclear.
   
     1) I am going to use the OPLS/AA force field, and as per the manual I will use the TIP4P water model, and the PME method.  However, when it comes to choices for rcoulomb, rlist and rvdw I am not clear.  The first OPLS/AA force field paper (Jorgensen et al 1996) that I can find doesn't have an explicit set of cutoffs, different cutoffs are specified for different organic liquids, and the PME section of the manual doesn't suggest values.  So what are the recommended values for rcoulomb, rlist and rvdw when using the OPLS/AA force field and PME calculation?  I've been using 1.0 for all three up to now.
     
     2) Since I am looking at short peptides, I am not sure what the best way is to define box size.  The extreme version is to assume a fully extended peptide backbone at 0.35 nm/residue and make the box diameter that length plus 2X the cutoff values.  However, that makes a generally excessively large box, since the fuly extended conformation is very unlikely to ever occur.  Is there some rule of thumb for specifying the smallest box that will avoid interactions between the instances of the peptide in different cells?
     
     Thanks for any and all advice.
     
     Warren Gallin