Hi all,
I am working on simulating a membrane protein with two different heme sites. I have parameterized following the MCPB.py tutorial and run a few minimization cycles. Upon inspection, after the first minimization step where all residues except water and ions are restrained with a 5.0 kcal/mol restraint, one of the heme groups gets very distorted causing massive issues when trying to heat (it also distorts even with larger restraint values). I believe there was maybe some error during parameterization of that heme group and generation of the frcmod file. The other heme group behaves as expected.
Going back through the tutorial I notice when generating the frcmod file for the heme using parmchk the note in the tutorial says they used parmchk instead of parmchk2 due to parmchk2 not generating enough parameters. I have used parmchk2 for my original parameterization and have now tried both to compare. The parmchk2 version provides the same number of parameters but with wildly different values and penalty scores of 136.0 for all dihedrals and one as high as 304 for an improper term. Parmchk does not indicate any penalty scores but perhaps this is due to the older nature of parmchk and maybe not recognizing its own error?
Does anyone have any insight on using parmchk2 vs parmchk for heme groups specifically or metal ligand sites in general?
I cannot figure out what else could be causing that distortion of the site other than bad force parameterization, but any insight on that would also be helpful.
Thank you!
Sarah
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Apr 07 2021 - 08:30:02 PDT
