Hi All,
We have been performing simulations on an NMR structure of a
calcium-binding protein, but the calcium ion does not remain coordinated
to the carbonyls in the binding site. At the beginning of the production
run, the protein conformation looks similar to the NMR state, but during
the simulation, the distances between the calcium ion and carbonyls vary
widely.
To run AMBER MD, we are using the CHARMM 36 FF, whose files were generated
with the CHARMM-GUI before using chamber to make the prmtop file. For the
calcium ion, we're setting epsilon=-0.120 and Rmin/2=1.367 according to
S. Marchand and B. Roux.
We are concerned about LJ parameters for the calcium ion, but they seem
accurate according to published references. Has anyone else run into
similar problems simulating calcium ions in proteins?
Caitlin Scott
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Nov 28 2014 - 16:30:03 PST
