VMD-L Mailing List

From: Philippe Bopp (philippebopp_at_REMOVE_yahoo.com)
Date: Mon Jul 22 2019 - 10:47:24 CDT

 Hi Maryam

please remember that VMD is just a graphics tool,  it knows nothing about "bonds",
just some criteria (usually distances) when to draw lines between atoms

>I have a system of 10 monomers in a simulation box trying to study aggregation and  I ran my simulation using gromacs. 
>However, after the simulation, I noticed that I have periodic boundary conditions i.e. some of my molecules were broken and I >also noticed some unwanted bonds/lines showing when I play the trajectory movie in VMD.
 
 This could have several reasons, among which

(I) your interactions potentials are no good and you have indeed in your configurations
    distances that VMD 'believes' (according to its criteria) to be 'bonds'

(II) the periodic box assumed by VMD (parallelepipedic suppose) is not the correct one,
   i.e. when particles are 'box-shifted' (e.g. by wrap/unwrap) too short distances occur

(III) probably more, eventually depending on input format (xyz, ......)

>I have tried to solve the issue using trjconv (gromacs) but its not fixed. 
>I also read through the Vmd mailing list and followed the instructions I found but the problem persists. Can someone please help?

   compute quickly some g(r) (with and without pbc) and see whether you find anything suspicious
 

>I will list the commands I have tried thus far below using vmd:
>Pbc unwrap (this removed the bonds and made the protein whole but the the protein went out of the box)

  this is OK, which coordinates are stored in a trajectory (eg. all in box, grouped by neutral entiries ....)
  depends on your simulation code

>Pbc wrap -compound -all (to move the proteins back inside the box), butv this did not completely solve the problem. 
>I have also tried 
>pbc wrap -center com -centersel "protein and chain A" -compound residue -all
>pbc wrap -center com -centersel "protein " -compound residue -all

 except for these 'bonds', which mean nothing, is there anything else suspicious in your configuration?

hope this helps

Philippe