[gmx-users] gmx spatial

[Christopher Neale]

Dear Hoda:

You’re likely seeing water mostly on one side because of PBC. I suggest you modify your commands such that you create a new .gro file in which the protein is centred and PBC is re-imaged (trjconv -center -pbc mol) and then use that to create a new .tpr file and then use that new .tpr file as input for the -s option in your protein fitting (while also using -pbc mol in that command)

I don’t know why you want to remove nonbonded water around the protein (Actually, I don’t know what is not nonbonded water…). If you mean you want the SDF of only the first hydration shell, then you can use gmx select to output particular water molecules and then (since I think you will need to have a contant number of waters) you can do some scripting to reset the coordinates of all other water molecules to 9 9 9 or something like that (though this may have -nab implications during gmx spatial).

Chris.

— original message —

I made Spatial distribution function of water around my enzyme, and I got a
grid.cube file, when I visualize my output file with VMD I see that water
distribution has tendency to one side of enzyme, it means when I change
isovalue in VMD the water molecules increase or decrease from one side of
enzyme but at last they cover whole the enzyme, here is two question:
1. are my commands and choices true?

echo 12 q | gmx_mpi make_ndx -f md_0_1.gro -o indexp.ndx   (12=water)

echo 0 0 | gmx_mpi trjconv -s md_0_1.tpr -f md_0_1.xtc -o md_0_1_S.xtc
-boxcenter tric -ur compact -pbc none (0=system)

echo 4 0 | gmx_mpi trjconv -s md_0_1.tpr -f md_0_1_S.xtc -o protein_fit.xtc
-fit rot+trans (4=backbone of protein , 0=system)

echo 12 1 | gmx_mpi spatial -s md_0_1.tpr -f protein_fit.xtc -n indexp.ndx
-nab 300 -bin 0.1 (12=water, 1=protein)

2. how can I remove nonbonded water aroun the protein?

I greatly appreciate your kind consideration and very much look forward to
hearing from you at your earliest convenience.

Best regards,

Hoda