[gmx-users] High energies after alignment of trajectory

[Hannah Magdalena Baumann]

Hi,

I want to calculate the free energy difference of turning some position restraints on. Therefore I run two MD simulations, one with the restraints on the protein turned on (I’m restraining the backbone of a specific loop), in the other one I have the restraints turned off. Then I rerun the trajectories of each state with the tpr of the other state and get the free energy difference using BAR. 
But since the protein undergoes rotational and translational movements during the simulations this does not just give me the position restraining energy. So I tried to align the trajectories before the rerun with
gmx trjconv -f prod_md.trr -o pbc.trr -s prod_md.tpr -pbc mol -center
gmx trjconv -f pbc.trr -fit rot+trans -o fit.trr -s prod_md.tpr
But evaluating the energies from this fitted trajectory gives me super large potential energies of the system (~e+12).
Is it generally possible to use a fitted trajectory to evaluate energies? 
Is there something wrong in the way I did the alignment?

Best,
Hannah