Shell solvation density

GROMACS version:
GROMACS modification: No

Hi,

I have a simple/general/dumb question, for clarification. I want to create a solvation shell around a solute at the standard water density, in a big box, that I will fill later with a different solvent.

If I run:
gmx solvate -cp solute.pdb -cs ../TIP3P_eq_box_6x6x6.gro -o solute_TIP3P_box.gro -p test.top -box 10 10 10

… the output is:

Output configuration contains 100518 atoms in 33506 residues
Volume                 :        1000 (nm^3)
Density                :     1002.34 (g/l)
Number of solvent molecules:  33505

So, the whole box is at 1 g/l. Now, if I run:
gmx solvate -cp solute.pdb -cs ../TIP3P_eq_box_6x6x6.gro -o solute_TIP3P_shell.gro -p test_shell.top -shell 1 -box 10 10 10

…the output is:

Output configuration contains 540 atoms in 180 residues
Volume                 :        1000 (nm^3)
Density                :     5.38472 (g/l)
Number of solvent molecules:    179

The density is higher, so I guess that the number is calculated with respect to the shell volume, not the box volume, right? (In order to get the solvation shell at 1 g/l I have to decrease the number of TIP3P waters with the -maxsol option.)

Thanks in advance.

No, the density is nearly a factor 200 lower in the latter case. The density is always computed with respect to the box volume

Thank you for the reply.

Which later case do you mean? When I use the -shell option or if I use the -maxsol option to decrease the number of TIP3P molecules? With -shell I get a higher density with less particles, that’s why I was wondering which volume was used for that calculation.

So, the -shell option, in general, will produce over-dense systems?

I don’t see any results with -maxsol in your post. With -shell the density is nearly a factor 200 lower.

This is what I get with the -maxsol option:

gmx solvate -cp solute.pdb -cs ../TIP3P_eq_box_6x6x6.gro -o solute_TIP3P_shell.gro -p test_shell.top -shell 1 -box 10 10 10 -maxsol 34

Output configuration contains 105 atoms in 35 residues
Volume                 :        1000 (nm^3)
Density                :     1.04703 (g/l)
Number of solvent molecules:     34

But have you now seen that the density is a factor 200 lower with -shell and even lower with -maxsol?

Oh sorry! I didn’t realize the decimal point in the first case, which gives basically kg/L.
Of course, with -shell the density is lower.

Thanks