From: Norman Geist (
Date: Mon Mar 02 2015 - 02:15:16 CST

Molid = Molecule ID (the number on the left in VMD main beside your molecule) e.g. 0

Centersel = The selection of your system for which the CenterOfMass is centered. e.g. “protein”



Norman Geist.


From: M K []
Sent: Monday, March 02, 2015 9:06 AM
To: Norman Geist
Cc: Josh Vermaas; VMD Mailing List
Subject: Re: vmd-l: Protein went out of the Box


Hi Norman


puts "Input) Molid?"

gets stdin mid

puts "Input) Centersel?"

gets stdin comseltext



Can you please tell me about mid and comseltext? what should be the input for these two?

I want to try your method, as I really need to get the system back to inside the box!




On Wed, Nov 12, 2014 at 5:15 PM, Norman Geist <> wrote:


Von: [] Im Auftrag von Josh Vermaas
Gesendet: Mittwoch, 12. November 2014 01:23
An: M K
Betreff: Re: vmd-l: Protein went out of the Box


Yup, because you told it to center the box on the moving protein, and only apply pbc corrections to the protein part, so PBCtools happily did nothing. What I think you want to do (assuming you have no interest in measuring the diffusion), is to move the entire system so that the protein's center is at the origin, and then rewrap it (if desired) so that the solvent molecules are on all sides of the polymer. The rmsdtrajectory tool has a nice interface for

O_o Oh careful here, you CANT use the rmsdtrajectorytool BEFORE wrapping as it does also rotate. Using an rmsd alignment destroys every relation to the periodic cell!!! A better way is to write a script that moves the center of mass to the cells center using only translations, afterwards just wrap.


puts "Input) Molid?"

gets stdin mid

puts "Input) Centersel?"

gets stdin comseltext


set frames [molinfo $mid get numframes]

set all [atomselect top all]

set comsel [atomselect top ($comseltext)]


for {set i 0} {$i < $frames} {incr i} {

  $all frame $i

  $all update

  $comsel frame $i

  $comsel update


  set com [measure center $comsel weight mass ]

  set lat [molinfo $mid get {a b c}]

  set shift [vecsub [vecscale $lat 0.5] $com]


  $all moveby $shift


puts "Info) Done!"




pbc wrap –all –compoud res



-Josh Vermaas

On 11/11/2014 06:12 PM, M K wrote:


I face a similar problem.

I simulate a polymer chain in water. I can select this chain using "protein" selection. During the simulation, it moves and goes beyond the boundaries of the box, and enters from the other side due to the periodic nature.

I want to keep this polymer chain always in the center of the cubic box; a good try is probably keeping the center of mass of the polymer chain in the center of the box. I tried the following code and some other variations, but they don't work; the chain is again split into parts when going beyond the boundaries:

pbc wrap -all -sel "protein" -center boundingbox -centersel "protein"

Can someone please help? What is the problem with the code I'm using?


On Tue, Sep 30, 2014 at 12:14 AM, Josh Vermaas <> wrote:

Hi Phan,

You can use the PBCtools plugin ( to rewrap your trajectory. If you just care about the hydrogen bonds between proteins, what you should be able to do is use something like this:

pbc wrap -all -sel "protein" -center com -centersel "some atomselection text that only picks out one of the two protein copies"

Then the selected protein will be near the center, and any of the atoms involved in hydrogen bonding should be inside the same unit cell. Read the documentation, and play around to see which combination of options you prefer.

-Josh Vermaas


On 09/29/2014 08:09 AM, Vy Phan wrote:

Dear Vmd users,

My systems for MD simulations included 2 protein, but after 50ns of trajectory, one protein go out side the water box (sometimes comeback). I knew why this phenomenon happened but when I do not know how to analyze the result.

I have problem when I calculated hydrogen bonds between two protein. Because of the boundary condition, the result of hydrogen bonds is not right. How can I solve this problem?

Thank a lot and best regards,

Phan Vy







Diese E-Mail ist frei von Viren und Malware, denn der avast! Antivirus <> Schutz ist aktiv.