From: Josh Vermaas (
Date: Mon Nov 23 2015 - 13:43:08 CST

Hi Bret,

PSF files aren't the only way of specifying bonds, just a convenient
one. I'm unfamiliar with VASP or how bonds are specified there, but in
principle you could specify all the bonds manually and write out a new
psf file for the structure. There is a command to explicitly set a
complete bondlist, but its a pain to use (see the atomselect interface).
A friendlier interface is found in topotools
(, which can be
used to add individual bonds between atomic indices. Once you've built
this internal datastructure once, you'd write it to a psf or other
structure-containing file. In a psf specifically, these would show up
under the BONDS heading.


On 11/23/15 10:55 AM, Bret Hess wrote:
> I'm using VMD with VASP output. I'd like to specify bonds in a PSF
> file, as I hear it's the only way to specify them in VMD.
> Since I'm not working with proteins, what skeletal content in the PSF
> file can I use to get VMD to read the bonds part?
> Thanks,
> Bret
> --
> Bret Hess
> Department of Physics and Astronomy
> Brigham Young University
> N251 ESC
> Provo, UT 84602
> 801-422-2108
> FAX 801-422-0553