Peptide not centered in box

User discussions
1

GROMACS version:
GROMACS modification: Yes/No
Here post your question

Hi
I am trying to run MD simulations on a peptide in gromacs (following an article called “information drivin, ensemble flexible docking using HADDOCK”)
for some reason when I do the MD my peptide escapes the box, I think it is because originally the peptide is not centered. (I see this in the file beta_solvated.gro)
showing my steps, what did I do wrong?

→ gmx_mpi pdb2gmx -f beta.pdb -o beta.gro -ignh -ff amber99sb-ildn -water tip3p
→ gmx_mpi editconf -f beta.gro -o beta_pbc.gro -c -bt dodecahedron -box 4.872
→ gmx_mpi grompp -f vacuum.mdp -c beta_pbc.gro -p topol.top -o beta_vac.tpr -maxwarn 1
→ gmx_mpi mdrun -v -deffnm beta_vac
→ gmx_mpi solvate -cp beta_pbc.gro -cs spc216.gro -p topol.top -o beta_solvated.gro

image
Thx!

2

My understanding is that editconf is using the center of mass of the entire system as the center, not just the protein. It looks like you have ions in the system (the purple spheres), if they were in the original pdb then editconf might have placed the center of mass of both those ions and the protein at the box center rather than centering just the protein. If you check beta_pbc.gro and this is the case, some suggestions are to either manually specify in editconf where the box is to be centered or remove the ions and add them back in later if they’re necessary to the simulation.

To be clear, due to periodic boundary conditions, the peptide moving out one side of the box effectively means it is moving back in from the opposite side, so if the concern is simply that the peptide will ‘escape’ the box and disappear, that wouldn’t be the case