On Sat, Aug 30, 2025, 徐文波 via AMBER wrote:
>
>I am conducting research on a CYP450 enzymatic reaction and have built
>a model of the key Compound I intermediate, which features an Fe(IV)=O
>species in the Heme center.
>
>The force field parameters for the metal center were developed using
>MCPB.py with the bonded model, based on quantum mechanics (QM)
>calculations. This approach worked perfectly for system building in tleap
>and for running stable production MD simulations in explicit solvent with
>AmberTools 24.
>
>Now, I am attempting to perform a binding free energy analysis using
>MMPBSA.py. The calculation fails at the beginning of the GB step with the
>classic "bad atom type" error. I have confirmed from my prmtop file that
>the atom type for the Fe(IV) ion is M1.
>
>
>My AmberTools 24 was install by Conda, and I didn’t find F90 files to add
>them into it.
>
>
>I understand that MCPB.py does not generate the GB radii required by
>sander, which is the source of this error. My question is about the best
>practice for this specific, advanced use case.
Sorry to be so slow in replying. The Amber GB models were never built for
metal ions, but in your case (with a "buried" Fe ion) things might be OK
But you will need to download the source code version of AmberTools, so that
you can modify the code in AmberTools/src/sander/mdread2.F90.
...good luck...dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Sep 12 2025 - 14:30:03 PDT
