AW: AW: question about wrap atoms to its original coordinates

From: Norman Geist (norman.geist_at_uni-greifswald.de)
Date: Fri May 20 2011 - 00:04:48 CDT

Hi,

yes, I know that problem of bond that seems to go through the whole box. To this first frame thing, I meant that if you use amber, you load your parmtop first and the original parmcrd as first frame before the dcd. On charmm it’s the psf and the pdb as first frame before the dcd. Also: If you use wrapAll you don't need to use wrapWater I guess, because it's already wrapped when you use "all". I think you just need to get into the unwrap command vmd provides. VMD is from the same people that wrote NAMD, so there should be a solution. Another thing is that vmd needs to know the cell dimensions. This is normally done when loading a parmcrd or pdb, but you can also set them by command in vmd.

The following should be exactly what you need, just try out ;)

set cell [options…]

Set the VMD unit cell properties (e.g. to use VMD’s feature that allows to display periodic copies of the system). 6

readxst xstfile [options…]

Read the VMD unit cell properties from an XST file. 7

get [options…]

Get the VMD unit cell properties.

wrap [options…]

When the atoms of the system are not all in one periodic image, but are distributed over various images, this function wraps all atoms into the chosen image. It is also possible to change between different representations of the unit cell (orthorhombic or triclinic).

unwrap [options…]

When overlong bonds (that stretch the whole system) occur and compounds (residues, segments, chains or fragments) are broken in the course of a simulation trajectory because atoms are wrapped around the periodic boundaries, this function will remove large jumps of atoms between consecutive frames.

join compound [options…]

When you have still broken compounds in frames after you have used unwrap, this function can be used to join broken compounds. Note, that this function is significantly slower than unwrap!

box [options…]

When you want to draw a box around the unit cell of your system, this function can be used. The box will automatically adapt to changes in the unit cell parameters in the course of a trajectory.

box_draw [options…]

When the unit cell parameters do not change in the course of a trajectory, this function draws a static box that will not adapt to changes in the unitcell properties.

Best regards

Norman Geist.

-----Ursprüngliche Nachricht-----
Von: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] Im Auftrag von jnsong
Gesendet: Freitag, 20. Mai 2011 04:32
An: Norman Geist
Cc: Namd Mailing List
Betreff: Re: AW: namd-l: question about wrap atoms to its original coordinates

Dear Norman Geist,

Thank you very much for help!

I'm so sorry for the late reply!

I have two question to ask for your help.

1.
In your MD simulation with NAMD, have you ever used "wrapAll on" and
"wrapWater on " in your NAMD input file?

2.
You have mentioned "Frame 0 is needed for reference and must contain the
original, unwrapped system." in your last letter.
I can't grasp the meaning of "the original, unwrapped system".
You know, I have used "wrapAll on" and "wrapWater on" in my simulation.
So, as long as I run MD simulation of my system, the first frame must be
wrapped by NAMD itself.
Do you mean that I should not use ""wrapAll on" and "wrapWater on" in
NAMD simulation?

I have tried "pbc unwrap -sel "resname RBF""(RBF is the name of my
ligand) and now the ligand and the protein are in the same image, but
the case is not true for the water.
It seems that the water have some wrong bonds after I typed "pbc unwrap
-sel "water"", because it looks awful in VMD OpenGL for water.

Hope your reply and help!

Thank your very much!

Jianing

 
On Wed, 2011-05-18 at 07:34 +0200, Norman Geist wrote:
> Hi,
>
> just load your system in vmd with a first frame where everything is there where it have to be, this is needed for reference, for example the crd or pdb file. Then load your trajectory and type the following to vmd console:
>
> pbc unwrap -first 0 -last last
>
> Frame 0 is needed for reference and must contain the original, unwrapped system.
>
> If this doesn't work, try with additional "-sel protein"
> If it is protein, if not, use another selection that fits your needs.
> This is what works for me most of the time, if it doesn't for you. Get into the "pbc unwrap" command.
>
> Best regards
>
> Norman Geist.
>
>
> -----Ursprüngliche Nachricht-----
> Von: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] Im Auftrag von jnsong
> Gesendet: Mittwoch, 18. Mai 2011 07:06
> An: namd-l_at_ks.uiuc.edu
> Betreff: namd-l: question about wrap atoms to its original coordinates
>
> Dear all,
>
> I have used NAMD to perform MD simulation on a membrane protein system.
>
> But in the trajectory, my membrane protein and its ligand are in
> different images, this brings me difficulty to analyze the simulation
> results. Take distance between an atom in ligand and another atom in
> protein for example, the value of distance will be very large if protein
> and its ligand are not in same image.
>
> Would you please to tell me how to solve wrap problem in NAMD?
>
> In my simulation, I have "wrapALL on" and "wrapWater on".
>
> Thank you very much!
>
> Jianing

-- 
Jianing Song
Ph. D. Student 
Institute of Theoretical and Computational Sciences
Dept. Phys.
East China Normal University
200062
3663 North Zhongshan Road
Shanghai, P. R. China

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:57:09 CST