By default MMPBSA.py uses a NAB program to calculate the GB potential. NAB
actually does not allow the internal dielectric constant to be changed (see
the epsext variable description on page 421 of the manual).
If you are so inclined, you can force MMPBSA.py to use sander (set
use_sander=1 in the &general section of the MMPBSA.py input file), then use
-make-mdins to generate the GB mdin file, add intdiel=2 or 4 to the &cntrl
namelist there (typically _MMPBSA_gb.mdin), then run again using the
"-use-mdins" file.
Note that GB has not been parametrized or tested with dielectric constants
besides 1.0 and 80.0 (or 78.5). Technically, the 'internal' dielectric
constant in GB really should be 1.0 --- it corresponds to the dielectric of
vacuum in the actual GB equation (the internal dielectric constant of the
protein actually cancels out in the deltaG expression).
The argument for modifying the internal dielectric constant as a model of
the dielectric of the interior of the protein is dubious, though not
indefensible. Trying to use any of the Amber GB models for anything except
water is probably flat-out wrong. PB is better suited for these types of
studies.
HTH,
Jason
On Wed, Jan 30, 2013 at 9:01 PM, <psu4.uic.edu> wrote:
> Dear Amber12,
>
> After reading Hou et al.'s "J. Chem. Inf. Model. 2011, 51, 69–82", which
> compares mmGBSA results with different *solute dielectric constant values
> (1,2,4)*, we are trying to perform a similar study for our own enzymatic
> system.
>
> The question is that Hou et al. used AMBER 9 "MM_PBSA", which allowed
> users to adjust solute dielectric constant values with the "intdiel" flag.
>
> However, in AMBER 12, we are unable to tune the solute dielectric constant
> values for mm*G*BSA in *mmPBSA.py.MPI*. We have tried tags such as "indi"
> , "intdiel", epsin". But the program will say "InputError: unknown
> variable ${X} in &gb" (${X}="indi" , "intdiel", epsin"). Could you kindly
> let us know if any equivalent tag in AMBER 12 *mmPBSA.py.MPI*? Thanks.* *
> *
> *
> Cheers,
> Henry
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Candidate
352-392-4032
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Jan 30 2013 - 19:30:02 PST
