[gmx-users] rigid body again

[Lianqing Zheng]

Dear gmxers,

I have checked the archive and it seems gmx is not directly supporting
rigid molecule simulation. I'm wondering if the following will work.

Say, for a molecule of N atoms, we choose 3N-6 independent internal
coordinates (e.g., from a Z-matrix) and constrain all these bonds (by
[constraints]), angles (by all-angles), and dihedrals (by adding 
improper dihedrals with large force constants). Will this achieve a rigid
molecule? Did anyone have any experience in this?

Thanks a lot for your help!

Lianqing