From: Peter Freddolino (
Date: Fri Jan 08 2016 - 13:57:47 CST

Dear Vlastimil,
>From the solvate manual page:

Beginning in VMD 1.8.6, it is possible to solvate a system with solvents other than water. In order to do this, you must prepare:
        • An equilibrated, cubic box of the solvent of choice (-spdb)
        • A psf file for the solvent cube (-spsf)
        • A topology file for all residue types in the solvent (-stop)
        • A key selection which will occur once in each solvent residue (eg., "name OH2") (-ks) **
        • The side length of the box (should be taken from the equilibration's periodic cell (-ts)
These five items are the **required** inputs for using an alternative solvent; if they are not specified (on the command line) or changed (in the GUI) standard TIP3P water will be used. To use an alternate solvent in the command line mode, **specify the five required inputs for solvate using the flags listed above**. **Please note that your solvent pdb must have segid QQQ for all atoms**. In addition, all solvent residues should be in the same chain and each residue should have a unique resid (ie, for mixed solvents, do not use separate residue numbering schemes for the multiple solvent types.

I’ve taken the liberty of starring several key sections of this documentation. We can tweak the command line help to make this clearer, but in general such help texts are a supplement, not replacement, for the real documentation.

Regarding the segname overflow, can you send an exact example so I can reproduce it?

> On Jan 6, 2016, at 4:27 AM, Vlastimil Zíma <> wrote:
> Hi everybody,
> I believe I found several bugs in solvate plugin when I tried to use it with custom solvent.
> * Custom solvent has to have a segment QQQ otherwise it doesn't work and it is very hard to find where the problem is.
> * All arguments for custom solvent are required, but than is neither stated not validated when using command line. This causes errors which are hard to debug.
> * Solvate plugin is able to generate invalid segnames, longer than 4 chars, if the is really large copies of the solvent.
> * Values for -ks argument are required to be selections, not the names of the atoms as suggested by the help.
> I hope at least some of the issues will be addressed.
> Vlastimil