GROMACS version:
GROMACS modification: Yes/No
Here post your question
I have a question for the community: for the same membrane+protein system in GROMACS with Martini, when using insane to add water and salt at -salt 0.15, I get NA: 2086, CL: 1106, W: 144943; whereas with gmx genion -neutral -conc 0.15, I get NA: 2965, CL: 1983, W: 146929. Which one is correct?
Hi there,
this is a delicate questions. Your systems seems to be highly charged (net charge approx. -1000e) you should be careful comparing actual ion concentrations with experiments.
Coming to your actual question: “Which one is correct?”
In my opinion, I would prefer the output of insane. gmx genion calculates ion concentrations based on the box volume, but a large fraction of the box volume is inaccessible to ions (membrane+protein). Your ion concentration in the bulk solvent would be therefore overestimated. insane should calculate it based on the amount of water in the system, excluding thereby effectively the volume taken up by the membrane and the protein. As a rule of thumb you can calculate ion concentrations with:
n(NA+) = 0.15/55.55 * n(Water) * X - 0.5 * q(Solute)
n(CL-) = 0.15/55.55 * n(Water) * X + 0.5 * q(Solute)
0.15 mol/l is the ion concentration and 55.55 mol/l the molarity of water at 298K. X is 1 for an atomistic force field and 4 for the MARTINI force field (one water bead in MARTINI equals to 4 “real” water molecules). In your case this results in: 1076 CL- & 2056 NA+ . So pretty close to the numbers that insane has provided. It deviates because I assumed 145000 initial water beads that may change and insane deviates a little bit from this scheme (if I remember correctly). Please take a look into insane.py if you want to know it exactly.
All the best,
Marius
1 Like