VMD-L Mailing List

From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Thu Jul 26 2007 - 07:41:24 CDT

Hi Stephan,
>
> Also I saw that "solvate" assumes only residues with name TIP3.
> solvate for non-standard should have an argument for non-standard
> resnames.
> Also the segment name "QQQ" is not a flexible thing. I made my copy
> of solvate.tcl and made changes therein.
Heh, thanks for noting the residue naming. I'll get it fixed in cvs. The
main case that I'd used this with was coarse-grained water, in which
case that didn't become an issue. The segname QQQ is indeed not
flexible, as documented at
http://www.ks.uiuc.edu/Research/vmd/plugins/solvate/.
>
> In autopsf-gui I saw that a LIG-patch must be specified last-first, i.e.,
> the first residue in the GUI-patch-editor must be the last of my chain
> and the second (the "optional one") must be the first of the chain.
Note that the autopsf gui is supposed to be general; the second residue
is listed as "optional" because some patches only require one residue
(eg, ASPP). It sounds like the design of this particular patch does want
the end of the chain first. Knowledge of the particular patches you're
using is necessary here.
>
> Any suggestions for building a DMSO-Box?
> I did it from the dmso.gro from Gromacs-Distribution and made it with
> explicit hydrogens due to Feller and Strader JPC(A) 2002
> with the following 2 files:
Everything here looks reasonable except for the topology entry; your
bond lines appear to have an odd number of atoms each, and most of the
bonds of the hydrogens don't appear to be specified (it's possible this
was mangled by email, but as far as I can see the lines end with the
third "C"). Also, the SD-OD bond isn't specified in any bond statements.
If you put it in a normal bond statement, the DOUBLE will be ignored.

Best,
Peter