From: John Stone (
Date: Fri Sep 17 2021 - 22:43:04 CDT

  VMD currently supports either STRIDE or DSSP to calculate secondary
structure. Both programs expect that the VMD molecul'es atom names/resid/etc
all correspond to the naming conventions used by the PDB. If you have
non-PDB atom names etc, then these programs will return an error and will
refuse to assign secondary structure for your molecule. You can use VMD
scripting commands to assign or rename any of the parts of your structure
that are missing the appropriate naming. This specific issue has been
discussed on VMD-L in the past, and I think Axel Kohlmeyer even wrote
some example scripts to show people how to efficiently do this renaming
at one point previously, since other LAMMPS users have had the same issues.

  John Stone

On Mon, Sep 13, 2021 at 06:36:43PM +0200, Runak R wrote:
> Dear VMD users,
> I have simulated a protein in LAMMPS. I am using the dcd file format as
> the output to plot the secondary structure in VMD. I got the
> segmentation error in VMD.
> I found that LAMMPS dcd output does not include the name of each segment
> of the protein (,and also the name of each molecule which is needed for
> secondary structure). Is there any alternative? If I can extract just the
> atomic properties (such as mass, atom type, atom ID, atom coordinates),
> not the residue properties, can I use the initial psf file of the protein
> that contains residue properties to create a trajectory file suitable for
> secondary structure? (using automatic psf builder) If no, is there any
> analysis that I can do with the protein alternatively?
> kind regards,

NIH Center for Macromolecular Modeling and Bioinformatics
Beckman Institute for Advanced Science and Technology
University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801           Phone: 217-244-3349