AMBER Archive (2004)

Subject: Re: AMBER: Bondi radii for BR and F in mm_pbsa calculation

From: Rhonda Torres (torres_at_scripps.edu)
Date: Tue Jul 20 2004 - 19:23:10 CDT

Hi Frank,

You are correct, Bondi radii are not present for those two atom types.
I would suggest you use the PBSA program in Amber to perform the molecular
surface calculation. With this, you will be able to use the radii from
your prmtop file.

# PROC - Determines which method is used for solving the PB equation.
# If PROC = 1, the delphi program is applied. If PROC = 2,
# the pbsa program of the AMBER suite is used.
# RADIOPT - Option to set up atomic cavity radii for molecular surface
# calculation and dielectric assignment. A value of 0 uses the cavity
# radii from the prmtop file. A value of 1 sets up optimized cavity
# radii at the pbsa initialization phase. The latter radii are
# optimized for model compounds of proteins only; use caution when
# applying these radii to nucleic acids.
#

If you would like to use the Delphi program to the
molsurf calculation, you will need to put them in the (Delphi) size file.

Hope this helps.

Rhonda

On Tue, 20 Jul 2004, S. Frank Yan wrote:

> Hi,
>
> We noticed an error message in the mm_pbsa calculations:
>
> Generate PQR
> No radius found for BR
> No radius found for F
>
>
> It was pinpointed to be in the mm_pbsa_calceneent.pm, in which the Bondi
> radii for different types of atoms are generated in the molsurf
> calculation:
>
> sub generate_pqr(){
> ###################
> ....
>
> # Bondi radii + 1.4A and probe radius of 0.0A yields SAS
> # Bondi radii + 0.0A and probe radius of 1.4A yields molecular surface
> # Bondi radii + 0.0A and probe radius of 0.0A yields vdW surface
> my %exp_rad = (
> "N" => 1.550 + 1.400,
> "H" => 1.200 + 1.400,
> "C" => 1.700 + 1.400,
> "O" => 1.500 + 1.400,
> "P" => 1.800 + 1.400,
> "S" => 1.800 + 1.400,
> "FE" => 1.300 + 1.400,
> "Na+" => 1.200 + 1.400,
> "Cl-" => 1.700 + 1.400,
> "MG" => 1.180 + 1.400,
> );
> ....
>
> However, the parameters for BR and F are not there. I was wondering if
> anyone has parameters for these elements which are compatible in this
> type of calculation.
>
> Thanks a lot,
> Frank
>
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
>

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu