[gmx-users] How to maintain the same exact positions of the protein and ligand in the extracted frame/pdb file as they were in the trajectory form?

[Mark Abraham]

On 4/06/2012 6:18 PM, Wayne Chen wrote:
> Hi,
>
> I'm a student at National Taiwan University doing research on protein-ligand
> interactions using MD simulations with GROMACS (version 4.5.3). In our
> simulations, we placed a protein (group A) and a ligand (group B) in a box
> under periodic boundary condition and ran for 20ns.
>
> I am trying to obtain snapshots of specific time frames within the
> trajectory file I got, but I am running into trouble outputting the exact
> position of the protein relative to the ligand when converting the frames
> into pdb files. Because of the period boundary condition, a snapshot may
> have the protein located on one side of the box and the ligand on the
> opposite end, but their actual relative positions---as measured by center of
> mass (COM) between specific region within the protein and the COM of the
> ligand----are in fact much closer to each other (1.33nm as calculated by
> GROMACS as part of the trajectory vs. 7.585nm after being converted into a
> pdb file).
>
> So my question is: How do I maintain the same exact positions of the protein
> and ligand in the extracted frame/pdb file as they were in the trajectory
> form?
>

See strategies here 
http://www.gromacs.org/Documentation/Terminology/Periodic_Boundary_Conditions

Mark