From: Peter Freddolino (
Date: Tue Jan 06 2009 - 11:51:45 CST

As the warnings that $sel writepsf emits would indicate, the psf written
in this way is not complete. You must instead use psfgen to apply your
patches (I would recommend making a patch for tryptophan to deprotonate
it, especially since using your method you aren't changing any of the
charges of other atoms in the Trp residue!) and write a psf through psfgen.
Peter wrote:
> Dear VMD experts,
> I have run a MD simulation with NAMD 2.6 of a protein in a water box and I
> obtained an equilibrated structure. Now I would like to use this structure
> (with hydrogens) as a starting point to do another MD simulation in which
> one of the amino acid residue (Trp) is deprotonated and the one next to it
> (Glu) is protonated. In order to do that, I made a glup patch to Glu
> residue and I wrote a new psf and a new pdb file. Then, I tried to
> eliminate the tryptophan proton using TkConsole:
> > set sel [atomselect top "not index x"]
> > $sel writepdb "newpdb".pdb
> > $sel writepsf "newpsf".psf
> Doing that, the psf and pdb files have the same numeration but the output
> of the MD simulation does not make sense (I can see atoms collapsing one
> upon each other!).
> Does anybody can help me?
> Thank you in advance for your suggestions,
> Caterina