Issue with re-wrapping and re-centering peptide-protein system

User discussions
1

GROMACS version: 2023
GROMACS modification: No

Hi everyone,

I’m running a 500ns MD simulation of 8 peptide-protein systems using GROMACS 2023 to investigate viral-host peptide-protein interaction. My protein (86 residues) is the same for all 8 systems, my peptides (8 residues) are different for each system, and each peptide-protein complex is solvated with TIP3P (no membranes or any other components included). I prepared all of my systems using CHARMM-GUI.
I have the complete 500ns simulations for all of my systems. My first issue was re-wrapping and re-centering my peptide-protein complex - I’ve read many posts from different forums on how to re-wrap and re-center protein-protein systems, especially when the chains split, causing several spikes/jumps in the RMSD/Radius of Gyration graphs. I managed to fix this issue for 6 of my 8 systems using the following “gmx trjconv” commands:

  • gmx trjconv -s md.tpr -f md.xtc -o md.whole.xtc -pbc whole
  • gmx trjconv -s md.tpr -f md.whole.xtc -o md.whole.nojump.xtc -pbc nojump
  • gmx trjconv -s md.tpr -f md.whole.nojump.xtc -o md.whole.nojump.mol.xtc -pbc mol -center -ur compact
  • gmx trjconv -s md.tpr -f md.whole.nojump.mol.xtc -o md_fit.xtc -fit rot+trans

However, for 2 of my systems, the spikes in RMSD/Rg graphs are still there, even after using these gmx trjconv commands. So, my question: can you please advise which gmx trjconv options (and in which order) are usually best for re-wrapping and re-centering the trajectories of peptide-protein complexes? Am I missing something with the gmx trjconv options I’m using, or maybe I’m using some of the options incorrectly?
I’ve tried several combinations of the options shown in the GROMACS2023 manual, but none of them seem to really work in the sense that I don’t get those spikes/jumps in RMSD/Rg graphs. I’ve also seen several forums speak about chains splitting due to periodic boundary conditions (PBCs), which is what I believe I managed to fix with the above commands. Please see attached an example of the a system and what I mean by “spikes/jumps” in RMSD, and a system where I “fixed” the spikes with the above commands.
Apologies for the long description. I’d really appreciate any help - I’m kind of at my wit’s end …

RMSD_spikes VS fixed
RMSD_spikes VS fixed1980×1125 154 KB

2

I faced the same issue.
After simulation just recenter the protein using the centring command
gmx trjconv -s md_0_100.tpr -f md_0_100.xtc -o MD_center.xtc -center -pbc mol -ur compact
1
0
I use this to center all my protein. If I face problem in RMSD I use this command
gmx trjconv -f MD_center.xtc -s md_0_100.tpr -o new.xtc -center -pbc mol
1
After that my rmsd problem was resolved and even RoG but for one complex my RoG showed spikes so I used this command and the problem was solved
gmx trjconv -f new.xtc -s md_0_100.tpr -o new1.xtc -pbc nojump
0
This does the job and I got my RoG without spikes
Best of luck

3

Hello,

I also faced same issue why simulating my protein and peptide complex using GROMACS. while doning anlaysis of simulated result i got spikes in my RMSD plot, and my complex is seperating with time, also my complex structure is getting out of the PBC. so I unwrap and remove the jumps and centered my complex in PBC box using following commands it works for me.

  1. Rebuild whole molecules (prevent split molecules):

gmx trjconv -s md_50ns.tpr -f md_50ns.xtc -o whole.xtc -pbc whole -ur compact

# When prompted: select group for output → choose “System” (whole system)

  1. Remove image jumps (continuous trajectories):

gmx trjconv -s md_50ns.tpr -f whole.xtc -o nojump.xtc -pbc nojump -ur compact

# When prompted: select output → choose “System”

  1. first create snashop.pdb , then check the chains id

gmx trjconv -s md_50ns.tpr -f md_50ns.xtc -o snapshot.pdb

choose : system

then create index.ndx with 2 chains : by entering chain A and then chain B

  1. gmx make_ndx -f snapshot.pdb -o index.ndx

enter : chain A

enter : chain B

enter : q for save and quit

  1. Center the slab/protein and rewrap around that cluster (keeps your slab in the middle):

gmx trjconv -s md_50ns.tpr -f nojump.xtc -o centered.xtc -pbc cluster -center -ur compact -n index.ndx

# Select group to cluster/center → choose your slab/protein group (e.g., “Protein”)

# Select group to write (output) → choose "System"

  1. (Optional) Ensure each molecule COM is inside the box:

gmx trjconv -s md_50ns.tpr -f centered.xtc -o final.xtc -pbc mol -center -ur compact -n index.ndx

# Select group to center → your slab/protein group

# Select group to write → “System”