Hi Jason,
So we also get the same error message recently by using ff12SB in
MMPBSA.py.MPI AmberTools 12/13. " PB Bomb in pb_aaradi(): No radius
assigned for atom 34 CB 2C"
We would love to change the F90 code in the vi
$AMBERHOME/AmberTools/src/pbsa/sa_driver.F90. However, since we might not
know the definition of "34 CB 2C" clearly here, we are not 100 % sure which
atom it is and where we should correct/modify in this "sa_driver.F90"
script.
In the $AMBERHOME/AmberTools/src/pbsa/sa_driver.F90,
#################################
! safe-guarding, very important!
do iatm = 1, natom
if ( isymbl(iatm)(1:1) == 'H' ) cycle
if ( radi(iatm) == ZERO ) then
write(6, *) 'PB Bomb in pb_aaradi(): No radius assigned for atom',
*iatm, igraph(iatm), isymbl(iatm)*
call mexit(6, 1)
endif
enddo
################################
May I know what *iatm, igraph(iatm), isymbl(iatm) *mean?*
*
* *When xleap generates this parameter file of which pb_aaradi encounters
the error, the corresponding pdb file is also generated. From this pdb
file, our wild guess is that this atom is the atom number 34 and it is the
beta carbon of the leucine (iatm = atom number), the atom type is amino
acid C beta. (atom type = igraph(iatm) ) *
*
ATOM 34 CB LEU 3 16.916 -12.796 -0.310 1.00 0.00
Here is the prmtop file of which pb_aaradi encounters this error:
%VERSION VERSION_STAMP = V0001.000 DATE = 09/25/13 20:13:44
%FLAG TITLE
%FORMAT(20a4)
default_name
%FLAG POINTERS
%FORMAT(10I8)
4005 17 2002 2042 4552 2768 8994 8937
0 0
22044 261 2042 2768 8937 104 233 271
58 0
0 0 0 0 0 0 0 0
71 0
0
%FLAG ATOM_NAME
%FORMAT(20a4)
N H1 H2 H3 CA HA2 HA3 C O N H CA HA CB HB2 HB3 CG CD1 HD1
CE1
HE1 CZ HZ CE2 HE2 CD2 HD2 C O N H CA HA *CB* HB2 HB3 CG HG
CD1 HD11
In the prmtop file, it does say this atom is an CB. Not pretty sure why
pb_aaradi still have issues to deal with this particular atom. Does this
error message suggest we should change this "CB" to "C2"?
We also try the other route by using parmed.py. In parmed.py,
either
change AMBER_ATOM_TYPE .%2C C
or
change AMBER_ATOM_TYPE .%CB C
or
change AMBER_ATOM_TYPE :LEU.CB C
or
change AMBER_ATOM_TYPE :LEU.CB 2C
All will bring the same error message: "CalcError: complex.prmtop!"
Could you kindly let us know if any solution?
Cheers,
Henry
On Tue, Aug 7, 2012 at 8:34 AM, Jason Swails <jason.swails.gmail.com> wrote:
> On Tue, Aug 7, 2012 at 9:03 AM, Vlad Cojocaru <
> vlad.cojocaru.mpi-muenster.mpg.de> wrote:
>
> > Dear Amber users,
> >
> > We are trying to use MMPBSA (AMBER12 + AmberTools 12) to estimate the
> > binding affinity of a protein to DNA. However, when we try to use the
> > non-linear Poisson Bolzmann (npbopt=1) with the the non-polar solvation
> > free energy estimated using the cavity and dispertion terms approach
> > (inp=2), we are getting the error "PB Bomb in pb_aaradi(): No radius
> > assigned for atom 14 CG C6" (see details below). The atom is a CG in a
> > ASP (C6 atom type in the ILDN-modified ff99SB force field). Please see
> > details of the run below.
> >
>
> The problem here is that the ILDN force field appears to not be frequently
> used, at least by those that work on PBSA (it's hard to keep up with all of
> the core atom types used by all of the force fields out there).
>
> There are 2 straightforward solutions to this issue.
>
> You can go through the sa_driver.F90 code (in
> $AMBERHOME/AmberTools/src/pbsa/) and add in recognition for the C6 atom
> type where you think it belongs (I think you may have to do this for
> multiple routines).
>
> Alternatively, you can use xparmed.py to change the atom type names from C6
> to C, which is the 'official' Amber force-field type for that particular
> atom (this will NOT change any parameters, just the name of the atom type).
> To take this approach, use the "change" command with the AMBER_ATOM_TYPE
> keyword (if you are using the GUI, you can use the arrow buttons to scroll
> to that key). The selection mask should be .%C6 to select all C6 atom
> types (or you can use :ASP.CG -- just make sure you capture all of the
> atoms you want to capture). Then set C as the new value.
>
> Now, if we use inp=2 and radiopt=0 (radii taken from the topology file;
> > we know we should not do this in principle - but we did it for testing),
> > the same error occurs.
> > However, if we use inp=1 and radiopt=0, MMPBSA runs but it gives
> > abnormal results (in the range of +4000 kcal/mol .. obviously a
> non-sense).
> > When we use inp=0 and radiopt=0, we get a better result (in the range of
> > -100 kcal/mol) but here we do not take into account the non-polar
> > solvation free energy
> > MMGBSA runs fine with the same system and gives reasonable values at
> > first glance (in the range of -200 kcal/mol).
> >
> > Could anybody explain this error message? Are we doing something wrong
> > here ?
> >
> > Also an additional question: how should we control the SASA calculation
> > in MMBSA ? ssaopt=0, saopt=0 are defaults. Does this mean that by
> > default no estimation of SASA is done since saopt=0? Does the gbsa
> > parameter play any role for MMPBSA or is used just for GB calculations ?
> >
>
> gbsa does nothing except for GB calculations.
>
> HTH,
> Jason
>
> --
> 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
>
--
Pin-Chih Su (Henry Su)
Ph.D. canditate
Center for Pharmaceutical Biotechnology (MC 870)
College of Pharmacy, University of Illinois at Chicago
900 South Ashland Avenue, Room 1052
Chicago, IL 60607-7173
office 312-996-5388
fax 312-413-9303
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Oct 11 2013 - 11:00:03 PDT
