Dear all
I have done multiple trajectories MD (100 ns in total or 5*20 ns) for my system with AMBER16. The Protein-ligand complex are mostly located in the edge of the water box during simulation and between 30-50 ns some protein residues (5-10 residues) move out from the water box. I have read some posts that it is because of PBC condition and imaging should modify correctly in VMD for PBC condition. I have tried to unwrap/wrap the protein and also shifting the center of the water box with "-shiftcenter {x y z}" and save in new .dcd file, now the protein-complex looks fine (stays in center) but the water does not have box-shaped and in some extend it splits/makes hole and move out from the protein surface. I would like to wrap it with CPPTRAJ and need to define center of mass with “anchor” argument in CPPTRAJ for protein dimer.
any help/suggestion for fixing the problem would be appreciated
Bests
Zahra
Hi,
the command you are looking for is either ”autoimage” or “image”.
In most cases autoimage will work nicely. It moves the center of mass of the first molecule (most likely you protein) to the center of the box and images the other molecules around it.
In some cases it will not produce the result you expect. If that’s your case, you should consider image. I usually use it in two steps but there are other ways to use it (see the Amber manual).
first set the center that you want with:
center [amber mask]
eg,
center :120@CA
to center around the alpha carbon of residue 120. Chose a relevant atom or set of atoms for your specific case.
Then run the image command:
image center
if you have a box that is a truncated octahedron, don’t forget to add the “familiar” option to the image command.
I hope it helps.
Good luck,
Antoine
Dear Antonie,
Hi.. thanks for your comment and suggestion.. I could fix the problem of wrapping and center in VMD but still have problem with autoimage keyword in RSMD calculation . now the box is stable during full simulation and protein-complex does not move out of the box. the issue now is my protein is in dimeric form and you can see in output file I have used for RMSD calculation, it takes only half part of the protein (monoer) as molecule one and/or anchor not the dimer.. now I am trying to figure out how to fix it... or it might not be important for RMSD result and its just warning about what CPPTAJ does not the error..
Cheers
Zahra
The warning just tells you that after running the RMSD command, you won't be able to use imaging because the box has been rotated in order to align all frames according to your mask and reference definition. If you just do RMSD with this input, then you can disregard this waring.
Best,
Antoine
Dear Zahra,
I would convert the trajectory into the pdb format and then reprocess the coordinates using a parser and a function that shifts the coordinates back to the center of the box. Using the box-dimensions (that information should actually be given in the input) you can shift any coordinate by that constant box-x,y,or z dimension. The pseudocode would look like that :
DO over all time-frames
parse coordinates over single time-frame
check if x,y,z are within threshold 0
Dear Emanuel,
Hi.. Thanks for your reply.. I could solve the problem with up-warp and re-wrap them .. the problem came from autoimage keyword which rotates the frame to align it into the reference structure for claculating. so it has been fixed by the following commands in VMD
pbc unwrap -sel "protein or resname SDF" -all
pbc wrap -centersel "protein or resname SDF" -center bb -compound residue -all
animate write dcd sdf_Wat_Sup.dcd
Cheers
Zahra
Hi, just to correct one point From Zahra’s last post. Autoimage does not rotate the frame, but the rmsd command does.
Best,
Antoine
Hi.. Yes Antoine.. I have used one script for merging trajectory+ rmsd calculation in that was the source of problem...
Cheers
Zahra
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Writing