I just want to give a feedback on what seems to have been a solution
to the problem I initially had with a Zinc containing protein
parameterized with MCPB. Here is the link to the other thread:
http://archive.ambermd.org/201207/0114.html
After adding the LJ parameters into the hydroxyl's hydrogen atom, I
was able to carry on with the simulation using pmemd.cuda without any
error messages or NaN energy values. The parameters used were the ones
found in the GLYCAM forcefield.
Although I had the simulation running, the resulting geometry of the
tetrahedral Zinc was still distorted. Apparently, the angle between
the ZN - O - H atoms collapsed to very small values (probably due to
the attraction caused by the sulfur atom linked to the zinc atom).
After another thorough read of the ZAFF paper
(
http://pubs.acs.org/doi/suppl/10.1021/ct1002626). When simulating
hydroxyl groups bonded to zinc, they also ran across the same problem.
The solution was to scale up the angle force constant. The value I
found on the additional material section was 23.82. I've been using
the same value and the tetrahedral seems to hold.
Thank you for the support.
Fabrício Bracht
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Jul 16 2012 - 13:30:03 PDT
