[gmx-users] How to center molecules in the box with trjconj ?

[ABEL Stephane 175950]

Hi all,
 
I have simulated a system containing 55 DPC molecules, a peptide some ions and water in a cubic box. Initially all molecules were placed randomly in the box. As expected during the simulation all the DPC molecules aggregate to form a stable micelle with the peptide located at the micelle interface. 
 
Now I would like to plot some MD configs in a PDB format some configs where the micelle and the peptide are placed in the center of the box. To take into account the PBC and to obtain the PDB, for the last frame at 150 ns as I used trjconv with the following commands: 
 
1-  trjconv_mpi -f traj_KTM17_DPC_all.xtc -s hMRP1_K-TM17_DPC.tpr -pbc nojump -b 150000 -e 150000 -o traj_KTM17_DPC_0ns_noJump.pdb
 
With "non-SOL" (DPC and peptide) as a group for output and where traj_KTM17_DPC_all.xtc is the trajectory with *no* PBC transformation 
 
2-  trjconv_mpi -f traj_KTM17_DPC_0ns_noJump.pdb -s hMRP1_K-TM17_dpc.gro -center -o traj_KTM17_DPC_0ns_noJump_center.pdb 
 
With DPC as a group for centering and "System" (DPC and peptide) as a group for output. the *.gro file contains only the DPC and peptide.
 
Unfortunately, these commands do not work since the micelle (and the peptide) are not in the box center. I have tried many others commands with -pbc as mol, res with no success. 
 
I used GMX version 4.0.7 
 
Any advices would be helpful.
 
Stefane