[Dock-fans] Combining DOCK with AMBER

Scott Brozell sbrozell at scripps.edu
Thu Nov 1 19:07:21 PDT 2007


On Wed, 31 Oct 2007, Francesco Pietra wrote:

> --- Scott Brozell <sbrozell at scripps.edu> wrote:
> 
> > On Tue, 30 Oct 2007, Francesco Pietra wrote:
> > 
> > > --- Scott Brozell <sbrozell at scripps.edu> wrote:
> > > 
> > > > On Mon, 29 Oct 2007, Francesco Pietra wrote:
> > > > 
> > > > > --- Scott Brozell <sbrozell at scripps.edu> wrote:
> > > > > 
> > > > > > On Mon, 29 Oct 2007, Francesco Pietra wrote:
> > > > > > 
> > > > > > > I posted previously that, with DOCK6.1, dock-prep for my protein,
> > after
> > > > > > > repeated fixing, arrived at fractional charges for the residues.
> > Either
> > > > > > some
> > > > > > > were still misnamed, or some were lacking. This was the best I
> > could
> > > > do.
> > > > > > > 
> > > > > > > Accepting the suggestion to go to Amber (someone was even skeptical
> > > > that
> > > > > > the
> > > > > > > fractional charges for this large receptor could be assigned by
> > > > > > Antechamber), I
> > > > > > > have now loaded the pdb to leap (Amber9), whereby heavy atoms were
> > > > added
> > > > > > (where
> > > > > > > I had set TER records) and H/lone pairs also added. Both prmtop and
> > > > inpcrd
> > > > > > > could be saved (and opened correctly with VMD). Of perhaps major
> > > > concern
> > > > > > > (leap.log attached):
> > > > > > > 
> > > > > > > (Residues lacking connect 0/ connect 1 -
> > > > > > > These don't have chain types marked:
> > > > > > > res  total affected
> > > > > > > CTHR 4
> > > > > > > NSER 4
> > > > > > 
> > > > > > These are innocuous.  They indicate that some residues are not
> > connected.
> > > > > > And those should correspond to your insertion of TER cards.  See
> > > > > > http://amber.ch.ic.ac.uk/archive/200502/0264.html
> > > > > > http://amber.ch.ic.ac.uk/archive/200505/0290.html
> > > > > 
> > > > > In fact, prepare_amber.pl lig.mol2 rec.pdb produced the expected files,
> > > > though
> > > > > 
> > > > > mpirun -np 4 dock6.mpi -i dock.in -o dock.out
> > > > > 
> > > > > after the "Initialing MPI Routines for all nodes, said
> > > > > 
> > > > > "getpdb: can't open file 0.amber.pdb"
> > > > > 
> > > > > The dock.out.# don't tell much.
> > > > 
> > > > 
> > > > http://dock.compbio.ucsf.edu/DOCK_6/6.1/bugfix.2
> > > 
> > > Ah! I failed to look at dock-fans. Sorry. However, the edited file
> > (attached
> > > base_mpi.cpp) gives the same error (getpdb: can't open file 0.amber.pdb) as
> > the
> > > original file (attached original_base_mpi.cpp).
> > > 
> > > Probably I was unable to follow the recipe, and there is no one around here
> > > capable of C or C++. I simply deleted a line of program from the original
> > file.
> > > 
> > > I have checked that the prmtop and inpcrd files generated by
> > prepare_amber.pl
> > > produce the correct representation in both VMD and Chimera.
> > 
> > 
> > base_mpi.cpp has been correctly patched.  Reinstall dock:
> > cd install; make dock
> 
> OK, thanks. However, reminding for installations where MPICH points to OpenMpi,
> to proceed as follows:


Yes, good point; bugfix 2 is for mpi.


> MPICH_HOME='/usr/loc' (or other appropriate location)
> export MPICH_HOME
> cd install
> make clean
> make dock # builds only the dock program
> 
> (otherwise errors arise about MPICH)
> 
> Now amber_score with option "ligand" 
> 
> mpirun -np 4 dock6.mpi -i dock.in -o dock.out
> 
> produced in a few minutes the expected files.
> 
> 
> > 
> > Scott
> > 
> > > > > This rec.pdb failed with Dock-prep in Chimera. Again, the program
> > believes
> > > > that
> > > > > there are non-standard residue, assigning them to Antechamber, which
> > > > produces
> > > > > fractional charges for the residues.
> > > > > 
> > > > > In other words, I did not succeed in getting Chimera and DOCK talking 
> > > > > together.
> > > > 
> > > > 
> > > > I probably do not understand exactly what you are doing,
> > > > why run Dock-prep again ?
> > > 
> > > It was to check if the pdb fixed by LEAP was now OK for DockPrep. It was
> > not.
> > > 
> > > Thanks
> > > francesco
> > > 
> > > > Perhaps you should open the amberized receptor files in Chimera:
> > > > rec.amber.pdb
> > > > rec.prmtop
> > > > etc
> 
> 
> That did not succeed (assuming that I kave implemented correctly the
> suggestion, which should be verified)
> 
> I created a directory, copying in the prmtop and inpcrd file for the receptor
> as obtained from LEAP (just the files that worked out perfectly for
> amber_score). Dock-prep (from Chimera, accepting defaults, i.e. everything
> selected except "Delete non complexed ions", and considering H-bonds, reported
> errors "Value error: underlying C++ Molecule object is missing", which is
> documented in the attacher error.log. 


Please post this small episode to Chimera-users.
Probably, this is an issue of operator error, but maybe not and
anyways it may help them improve robustness.


> However, the error is different from that obtained with the pdb file created
> from the above prmtop and inpcrd using "ambpdb" of Amber9. In the latter case,
> apparently, some standard amino acid was erroneously taken by Dock-Prep as
> non-standard residue and passed to Antechamber, which failed to do correctly,
> reporting fractional charges for the residue. For the present protein, the
> relevant window that was presented, was
> 
> ASH+ALA+ASN  +0
> ASH+ALA+PRO +0
> ASH+GLY+MET +0
> GLH+ALA++ALA +0
> GLH+ARG++LEU +1
> GLH+ARG+THR +0
> 
> accepting that, fractional charges were calculated for the residues. Any change
> to the values (+0 or +1) presented led to crashing.
> 
> What I was trying was to run some simple docking before amber_score. The


This is mandatory since Amber_score cannot be used for docking,
only for rescoring (as well as for serious MD but as indicated before
it is better to use the Amber package itself for serious MD).

> tutorial passed without problems. Even my protein passes correctly Dock-prep,
> if prepare from Chimera is accepted, and I could run all dockings as in the
> tutorial. However, the protein handled by Chimera is not accepted by Amber9 or
> amber_score.


This is not surprising since Amber's LEaP (which is the vehicle for
amberization under the hood of DOCK's prepare_amber.pl)
is the sole mechanism for Amber input preparation.


Here is the usual sequence:
Do docking (along the lines of tutorials 
Structure Preparation,
Sphere Generation and Selection,
Grid Generation,
Docking );
Then rescore with Amber_score (along the lines of tutorial
Structure Preparation & AMBER Score ).
Note that Amber_score has receptor input files distinct 
from DOCKing receptor input files.

Typically, one has an original pdb which is the starting point for 
dock prep, and the resulting mol2 file (from step 2f in the
Structure Preparation tutorial) also can be saved as a pdb;
it is this later pdb that is usually the starting point for
prepare_amber.  However, one can imagine several ways to
obtain the actual receptor input files; for example, one might
take the pdb file from step 2g and use that for prepare_amber.
This would be a simple consistency test for protonation
between the DOCKing receptor and the Amber_score receptor.

It seems to me that you are performing various consistency checks
between DOCKing inputs and Amber_score inputs.
That is a sound approach, but human inspection will play a 
dominant role since the DOCK and the Amber packages are
not input compatible (despite the appearance through the
magic of amberization).

[Note to Amber/DOCK superexperts: there are several caveats
that I have ignored for pedagogical purposes.]

Scott

> > > > Scott
> > > > 
> > > > > > > Should all that be OK, how to go to DOCK now? I have prmtop and
> > inpcrd
> > > > > > files
> > > > > > > also for the ligand, obtained from prepare_amber.pl, though,
> > obviously,
> > > > no
> > > > > > such
> > > > > > > files for the complex receptor-ligand.
> > > > > > 
> > > > > > If prepare_amber.pl was run, there should be all the Amber files
> > (prmtop,
> > > > > > etc)
> > > > > > for all three (ligand, receptor, complex).
> > > > > > See step 3 in
> > > > > >
> > http://dock.compbio.ucsf.edu/DOCK_6/tutorials/amber_score/amber_score.htm
> > > > > > Once amberization is complete and verified then use step 4 of that
> > > > tutorial
> > > > > > as a template for creating your own dock input file for rescoring.
> > > > > > 
> > > > > > > Also, the receptor has 24.000000 charge: should docking be carried
> > out
> > > > with
> > > > > > > this charged receptor or should it be first neutralized (in leap?).
> > > > > > 
> > > > > > DOCK's Amber_score uses an implicit solvent model in which case
> > > > > > charge neutralization with explicit ions is usually not performed.



More information about the Dock-fans mailing list