On Wed, Sep 7, 2011 at 1:28 PM, Peter Varnai <P.Varnai.sussex.ac.uk> wrote:
> Many thanks for the responses.
>
> Jason, do you know what went wrong in using the python script?
>
Yea, the script is old and not well-versioned. I think I had added a hack
into the chemistry package to allow this program to run without specifying a
coordinate file, but that must have been after AmberTools 1.5 was released
(that and all of my development efforts in this direction have been put
towards parmed -- I actually had to pull this script out of the depths of
the git logs). The easiest way to change it was to just require a
restart/inpcrd file. Try the attached one instead.
>
> [zeus]$ $AMBERHOME/bin/python ChangeParmRadii.py -p peptide.top -r mbondi3
> -n peptide3-mbondi3.top
> Reading in your topology file peptide.top
> Changing radius set to mbondi3
> Traceback (most recent call last):
> File "ChangeParmRadii.py", line 217, in <module>
> mbondi3(parm)
> File "ChangeParmRadii.py", line 144, in mbondi3
> mbondi2(parm) # start from mbondi2 radii
> File "ChangeParmRadii.py", line 114, in mbondi2
> molecule = parm.ToMolecule()
> File
> "/cvos/shared/apps/amber/amber11/lib/python2.6/chemistry/amber/readparm.py",
> line 883, in ToMolecule
> if self.valid and self.rst7.valid:
> AttributeError: amberParm instance has no attribute 'rst7'
>
> cheers,
> Peter
>
> On 7 Sep 2011, at 06:36, Jason Swails wrote:
>
> > On Tue, Sep 6, 2011 at 2:38 PM, Carlos Simmerling <
> > carlos.simmerling.gmail.com> wrote:
> >
> >> the manuscript on these is under review. I think Jason's script is
> ready.
> >> mbondi2 radii are ok but mbondi3 are optimized and do indeed give much
> >> improved salt bridge strength and other properties.
> >>
> >
> > I had fixed the issues my script had previously (which prevented me from
> > including it in the AmberTools 1.5 release). The script was a
> > quick-and-dirty solution to changing radii sets and is *only* capable of
> > setting intrinsic radii to one of the leap-specific sets (bondi, mbondi,
> > mbondi2, amber6) and mbondi3 (as described in the paper Carlos is
> > referencing), and I've attached it here.
> >
> > Quite awhile ago I actually incorporated the functionality found in this
> > program (along with all of my other little topology file editing scripts)
> > into a single parameter file editor that is vastly more flexible than the
> > script I included here which should make many tasks easier (and a couple
> > others possible in the first place). This will be included (I
> think/hope)
> > with the next AmberTools release, so I'm just tossing in a little advance
> > advertisement here :).
> >
> > To run the script I attached here, you'll need to have AmberTools 1.5
> > installed. This script depends on the chemistry python package directory
> > found in $AMBERHOME/AmberTools/src/etc/chemistry. As a result, it's
> easiest
> > if you allow AmberTools 1.5 to build its own version of python. If you
> do
> > this, then you can execute the script included as follows:
> >
> > $AMBERHOME/bin/python ChangeParmRadii.py --help
> >
> > That should tell you how to use it.
> >
> > Hope this helps,
> > Jason
> >
> >
> >> On Tue, Sep 6, 2011 at 1:56 PM, Peter Varnai <P.Varnai.sussex.ac.uk>
> >> wrote:
> >>
> >>> Dear All,
> >>>
> >>> This is to inquire whether there has been any update regarding the
> mbondi3
> >>> parameters for protein simulations with igb=8.
> >>>
> >>> The amber11 manual states (p. 57) that the mbondi2 radii are fine to
> use
> >>> with igb=8 which Carlos mentioned is the case indeed on 27 April but
> some
> >>> new modifications are available to improve those results for salt
> bridges.
> >>> The manual also details new scaling and atom-specific parameters which
> all
> >>> seems to have been implemented in mdread.f (apparently these can be
> changed
> >>> in the mdin file).
> >>>
> >>> I would be happy to update the protein atom radii in the prmtop file to
> the
> >>> most recent mbondi3 values: (i) has the paper been published that gives
> >>> those values? (ii) Jason's python script may be made available that
> corrects
> >>> for those? (iii) is there a leap fix available?
> >>>
> >>> Many thanks,
> >>> Peter
>
>
> _______________________________________________
> 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 Sep 07 2011 - 13:00:03 PDT
