VMD-L Mailing List

From: Ming-Tsung Lee (mtlee_at_rutgers.edu)
Date: Mon Aug 01 2016 - 17:55:37 CDT

Dear VMD users,

Problem: My system contains 100 molecules of 10 coarse grained beads with
two bead types. The structure of the molecule is C-C-C-C-C-N-N-N-N-N, which
contains 9 bonds for each molecule. Box size is 160 in all dimensions, and
the box is centered at origin. The trajectory is in PDB format, which
contains the atom coordinate and the connectivity. The follows is the
download link for an example of 20 frames.

trajectory (1.8 MB)
https://www.dropbox.com/s/34oy86zuto5v1qe/frame8.pdb?dl=0

If I load it in PDB, I will see some over-long bond, e.g. on frame 8 (7 on
VMD) or the last frame.

I fail to debug the problem, so I am seeking for some hint from more
experienced users. My problem is not similar to the questions posted on the
forum about the overlong bonds on the opposite sides of the periodic
boundary. Yet, I have tried the suggestions by using pbc unwrap. The
command fixed some frames, but makes other frames worst.

Some of my attempts may help you to resolve the problem (or confuse you
more):

1. The overlong bonds only happen on atom name N.
2. If I output a single "problem frame" into a new PDB file, the problem is
gone. For example, the following link contains the 8th frame:

frame 8 (76 KB):
https://www.dropbox.com/s/erw977vgqhfjv1b/over-long.PDB?dl=0

Any help is appreciated. My apologies if the problem is due to my
misunderstanding of VMD function. FYI, my VMD version is the latest for MS
Windows (1.9.3).

Best regards,
Ming-Tsung 

--
Ming-Tsung Lee (李旻璁)
Postdoctoral Researcher
Dept. of Chem. Eng. | National Taiwan University
+ 886 919 050 488 (Taiwan)