From: Nicola Giacche' (
Date: Thu Mar 05 2009 - 09:01:08 CST

Hi Brittany,
the problem is that when I use the pbc unwrap command in the whole
trajectory, the results is a "broken" system with wrong bounds etc...
then I would apply the command only to select frames... and for a
trajectory of 5ns I can manually select the interval of "jumping" frames
(I have about twenty intervals to select), but for longer trajectory I
would like to use an automatic command or a script.
If there is a command in vmd able to recognize, in the trajectory, the
jumping frames it will be perfect for me. I will make a script with
first the recognition of frames and then I will apply the pbc unwrap
command only to the selected jumping frames.
Does exist something like this?


Hi Nicola,

I am not sure. The way pcb unwrap works is that it bases it on the previous
frame - if an atom moves more than a certain amount between frames, it
assumes it has jumped. So if you wanted to avoid processing all of the
frames, you could split up the trajectory into pieces with one normal frame
before each of the jumping frames and process them individually. I am
guessing that system size is the reason you don't want to process the whole
trajectory together?


-----Original Message-----
From: Nicola Giacche' []
Sent: Thursday, March 05, 2009 5:41 AM
To: Brittany Morgan
Subject: Re: vmd-l: script to resolve trajectory "jumps" on Desmond dynamic

thank you Brittany, the command pbc unwrap -sel protein works, but now I
have the problem to select only the frames that have "jumps"... is there
in vmd an automatically command that select the "jumping" frames?
for example: I know the coordinate limit value... then all frames with
an atom over this value jump...

thank you again


Brittany Morgan ha scritto:
> Hi Nicola,
> I use 'pbc unwrap'.
> Brittany
> -----Original Message-----
> From: [] On Behalf
> Nicola Giacche'
> Sent: Wednesday, March 04, 2009 11:15 AM
> To: VMD Mailing-List
> Subject: vmd-l: script to resolve trajectory "jumps" on Desmond dynamic
> Hi,
> I would to resolve a problem of Desmond dynamic... in some frames of the
> dynamic, when an atom of the protein goes beyond the water-box, the
> protein "jumps" to the next water-box... precisely, if the atom goes
> beyond the Y direction, the component Y of the coordinates of the
> protein "jumps" of value of the lenght Y of the water-box.
> Is there a script on VMD able to restore the correct Y coordinates of
> the protein (or chain), automatically, in the frames with the "jumps"?
> thanks in advance,
> Nicola

Nicola Giacchè,
PhD Student
Bachelor Degree in Medicinal Chemistry and Technology of Drugs
University of Perugia
Department of Medicinal Chemistry and Technology of Drugs
Via del Liceo, 1 - 06123 Perugia (Italy)
Tel ++39 075 585 5169/5156