I am working on membrane-peptide interaction. I’m planning to perform an
umbrella sampling simulation to find out the thermodynamics of the
antimicrobial peptide-membrane binding in Gromacs-21 software.
Initially, I placed the peptide 10.7 nm away from the center of mass of
the membrane. Now I want to pull the peptide up to the membrane center
(the center of mass distance is zero). I’m struggling in the pulling
part. I’m pasting here my pulling commands. It will be a great help if suggest what are the parameters I should add in my mdp file.
pull = yes
pull_ncoords = 1 ; only one reaction coordinate
pull_ngroups = 2 ; two groups defining one reaction
coordinate
pull_group1_name = peptide
pull_group2_name = membrane
pull_coord1_type = umbrella ; harmonic potential
pull_coord1_geometry = distance ; simple distance increase
pull_coord1_dim = N N Y
pull_coord1_groups = 1 2
pull_coord1_rate = -0.15
pull_coord1_start = no ; define initial COM distance > 0
pull_coord1-init = 0.0 ; define initial COM distance > 0
pull_coord1_k = 6000 ; kJ mol^-1 nm^-2
pull-nstxout = 50 ; kJ mol^-1 nm^-2
pull-nstfout = 0 ; kJ mol^-1 nm^-2
I guess you are using this simulation to generate starting configurations for umbrella sampling, right? There are a few things I would suggest changing in your pull settings:
pull_coord1_start = yes ; define initial COM distance > 0
With your current settings (“pull_coord1_start = yes”, “pull_coord1-init = 0.0”) the first thing that you will do is that you will instantly pull the peptide to the center of the membrane and from there you will reduce the distance (which must not be negative). There would be a huge force in the first step and then a crash (either due to instabilities or the negative pull distance).
You also pull with a very high pull rate. Starting at 10.7 nm distance you can pull for 71.333 ps before you end up with a negative distance. I would suggest:
pull_coord1_rate = -0.001
nsteps = 5350000 (if you use dt = 0.002)
Then you will reach 0 distance by the end of you simulation. It might crash very close to the end if your initial distance was slightly lower than 10.7 nm. But you will still get the configurations you are interested in. You will also need the simulation box to be large enough, so that the initial distance of 10.7 is < 0.49 * [box Z dimension].
First, make sure that you are not applying any restraints on the peptides, by mistake.
Since you are pulling the center of mass of multiple molecules, applying the same pull forces to all atoms in the group, it might end up easier to just pull some of the molecules, if they are loosely bound to each other, while the others remain behind.
It might be worth trying keeping them in two separate selection groups and making two separate pull coordinates, one for the peptide and one for the nanocluster. But you might need quite high pull force constants so that they remain together. Hopefully that will help.
Another alternative worth trying, is to keep the first pull coordinate as it is and add a second one pulling the COM of the lipids to the COM of the nanocluster, using pull-coord2-start = yes and pull-coord1-rate = 0 so that they keep the same relative COM distance.
Sorry. Basically the peptides are attached to the nanocluster through sulpher atoms. It is peptide functionalized nanocluster. Now I’m pulling the nanocluster by removing all position restrain. It works! Thanks Sir. Sorry for my misinformation.