You have to consider the periodic boundary conditions to calculate the accurate distances. I can send you a script for VMD, but am at the phone right now.

Please find attached a TCL script to run in VMD (source in Tk Console) that calculates the COM-COM distances with PBC for all frames and stores them into a simple text file that can be plotted with e.g. XmGrace

Norman Geist Thank you sir for providing the script.

I tried to use the script with two cases. first when the PBC is not removed and the second with PBC removed and they produced different values.

Which case is the correct one?

Thanks

when i examined my trajectory, i found that the system breaks due to the PBC, so i removed the PBC in vmd using this command:

pbc unwrap -sel "protein or resname UNK"

The script computes the centers of both selections. For this stage, the PBC isn't considered so these "fragments" should be whole. For the distance between both selections, the PBC is considered and it shouldn't make a difference if you have a wrapped or unwrapped trajectory, as long as your molecules are whole.

Unwrapping a system can be really tricky and that's why I usually dont generate wrapped trajectories at all (NAMD). To get your "fragments" whole, you can try something like:

#join all fragments in first frame

animate goto 0

pbc join connected

#now unwrap all frames based on the fragments in the first frame

pbc unwrap

PS:

#for future wrappings consider: (keep fragments whole)

pbc wrap -all -compound fragment

@Norman Geist

when i used this command "pbc join connected" after animate goto 0, some atoms were stretched too much. However, when i used this command (pbc unwrap -sel "protein or resname UNK") only the protein became whole again with no errors. so should i stick to this command to unwrap my trajectory before making any analysis?

and sorry for asking too much

Ibrahim M. Ibrahim if both your selection are then whole over the course of the trajectory, the script should work just fine. You should stick with whatever combination of wrap/unwrap works for you.

In case you are using NAMD for MD, leave all wrap commands out for the future.

Bests

@Norman Geist

Thank you sir for your help.

Sorry, I did not understand you well in the last comment.

"In case you are using NAMD for MD, leave all wrap commands out for the future."

Could you please describe what you meant by leaving all wrap commands out for the future?

Thanks

I meant leave out commands like wrapAll, wrapWater wrapNearest in the NAMD input.

Norman Geist I checked my input file and found these

wrapWater on; # wrap water to central cell

wrapAll on; # wrap other molecules too

wrapNearest off; # use for non-rectangular cells (wrap to the nearest image)

does this mean that the run was incorrect?

No it wasn't incorrect, it just creates wrapped trajectories and this leaves you with this necessity to unwrap again. I find it much more comfortable to generate unwrapped trajectories and wrap them in VMD When needed.

From your last answer I get that there is a difference between the wrapping the trajectories and applying PBC.

could you please tell me what is the difference?

Yes, you can treat the system periodic durung the simulation without wrapping the coordinates for the trajectory output.

I did not know that. Thanks for telling me

Norman Geist

i have run a trial MDS of a protein ligand complex by setting theses commands to off (wrapWater, wrapAll, wrapNearest)

the first image is the first frame of the trajectory:

and the second image is the 900th frame:

is this correct?

Yes, you see now continues diffusion of the all atoms, especially water. You can still wrap the coordinates if needed and this is much simpler than unwrapping. Due to the diffusion, the coordinates will get too large for file formats like PDB at some point. However, this usually only occurs after several microseconds.

Thank you sir for your help