Hello
I am running protein-ligand complex simulations- with protein embedded in lipid bilayer and solvated + ionized as per standard rules of Gromacs.
After complete Energy minimization - in NVT step the bilayer starts separating from as shown in figures attached.
Tried using different combination of position restraints and without position restraints as well - but still bilayer separates.
Not clear what casing this ambiguous behavior in lipid bilayer.
NVT.mdp parameters
define = -DPOSRES ; position restrain the protein and ligand
energygrps = Protein AGO
tc-grps = Protein_AGO Water_and_ions POPC
tau_t = 0.1 0.1 0.1 ; time constant, in ps
ref_t = 300 300 300 ; reference temperature, one for each group, in K
Leaflet separation is common in NVT simulations. Either restrain the P atoms in the z-dimension or just run a long NPT equilibration to let the box compress back and the leaflets will close down.
Thanks Justin for the reply.
Just to make sure P atoms -- Phosphorous atoms of lipid molecules.
Because when I try to generate position restraints for POPC in Z-dimension. They are generated for all the 52 atoms of the POPC molecules from:
gmx genrestr -f sys_solv_ion.gro -o popcZ_res.itp -fc 0 0 1000
>>>>
[ position_restraints ]
; i funct fcx fcy fcz
3101 1 0 0 1000
3102 1 0 0 1000
3103 1 0 0 1000
3104 1 0 0 1000
--------------------------------------------------------------
regards
Ajay
Position restraints are assigned on a per-moleculetype basis. You can't pass the full system to genrestr and get a sensible result, as noted in the help description of the program (and many online resources). Write a single-atom [position_restraints] directive by hand; it's much easier.
Hi Justin, a way easy to generate a membrane with a protein encrusted is using the server http://www.charmm-gui.org/?doc=input/membrane , this Membrane Builder helps the user generate a series of inputs (files) necessary to build a protein/membrane complex for molecular dynamics simulations in different software like gromacs, amber, charmm or namd, among others. I suggest you generate the file and compare with you will generate. I've run a simulations around 90 ns without this troubles.
- Badges
- Science topic