[gmx-users] Restraining one water molecule - settles vs constraints

[Hermann, Johannes]

Dear Gromacs Users, dear Justin,

I am using the tip3p water model and I want to restrain one single water 
molecule. I found Justins reply in the mailing list a few years ago:

>>/If you want to restrain a single water molecule, it needs to be defined as />>/its own [moleculetype] or as a part of the protein [moleculetype]. 
Breaking />>/apart a continuous block of water causes problems with the SETTLE />>/algorithm, so you will need to manually specify three constraints (OW-HW1, />>/OW-HW2, and HW1-HW2) for this water molecule, while the remaining bath of />>/solvent would be handled by SETTLE. />>//>>/-Justin /

I looked up in the manual how the constraints-section is formated, but I 
have no glue about the funct type. Perhaps funct 1? For the lengths of 
OW-HW1 and OW-HW2 I would assume that I take the bond-lengths specified 
in tip3p.itp, but I have no idea about the HW1-HW2 length.

[ constraints ]
; i   j funct   length
1     2  1  0.09572
1     3  1  0.09572
2     3  1  ??

Can anyone help me?

Thank you very much!

All the best
Johannes