[gmx-users] geometry is rigid, but angle energy changes

[Oleg V. Stroganov]

Dear gmx-users!

I have observed strange behavior of energy in my simulation. I use
oplsaa and gromacs 3.2.0 to obtain molecular-mechanics energy profile
of aminogroup rotation. I have constructed trajectory in which the
whole geometry of ligand remains unchanged, except rotation of NH2
group around N-CA axis. -dummy hydrogens option of pdb2gmx was
implemented to describe amine. Contrary to my expectations not
only non-bonded contributions changed during md-rerun of artificial
trajectory, but also angle energy oscillated harmonically with
amplitude of ~7 kJ/mol.

When I turned dummy hydrogens off, these odd oscillations of angle
energy disappeared.

How can i deal with this phenomena? Is there any way to get rid of
undesirable oscillations of angle energy for rigid geometry?


-- 
Best regards,
 Oleg Stroganov, Moscow State University.