You have to re-install MMPBSA.py in order to import the modules with the
correct python version. Another option is just doing this by hand:
cd $AMBERHOME/src/mmpbsa_py
python2.7 -c "import alamdcrd inputparse utils"
mv *.pyc ../../bin
Good luck!
Jason
On Tue, Feb 22, 2011 at 11:16 AM, <poll.chemie.uni-hamburg.de> wrote:
> Hi Jason,
>
> Thank you for these background informations. I installed python2.7 and
> tried the MMPBSA.py again and still got the same error. Maybe it could
> help to re-install the MMPBSA.py stuff?I'm using Amber11 so MMPBSA.py
> is installed by default but maybe this went wrong.
>
> Quoting Jason Swails <jason.swails.gmail.com>:
>
> > Hello,
> >
> > Python 3.0 is basically a re-design of the python language. As
> > such, code
> > written in python2.x cannot be directly run with python3.x
> > interpreter.
> > There are many websites that detail this.
> > http://wiki.python.org/moin/Python2orPython3 is one of them. For
> > MMPBSA.py,
> > you need to use either python2.6 or python2.7 (though this is
> > changing in
> > the upcoming release). In order to get it to work for python3.x,
> > you have
> > to pass each module and program through the "2to3" utility. Type
> > "which
> > 2to3" to verify that you have this program.
> >
> > However, you have an error that pops up before this becomes an
> > issue, I made
> > a comment below.
> >
> > On Tue, Feb 22, 2011 at 9:26 AM, <poll.chemie.uni-hamburg.de>
> > wrote:
> >
> > > I tried to use MMPBSA.py.As mentioned in the manual a python
> > version
> > > 2.6 or later is needed. I used python3.0 which is already
> > installed
> > > on our machines. As I executed the MMPBSA.py I got the following
> > > error:
> > >
> > > python3.0 $AMBERHOME/bin/MMPBSA.py
> > >
> > > Traceback (most recent call last):
> > > File "/opt/amber11/bin/MMPBSA.py", line 44, in <module>
> > > import inputparse, utils, alamdcrd # (2)
> > > ImportError: Bad magic number in /opt/amber11/bin/inputparse.pyc
> > >
> >
> > Just a little background to make this error make some sense: The
> > first time
> > a module is imported by python, the python interpreter "compiles"
> > the module
> > into what is called "byte-code", which is a binary .pyc file that
> > loads
> > faster than a text-based module. However, each version of python
> > creates
> > *different* byte-code binaries, so a .pyc file created with one
> > version of
> > python (i.e. python2.5) cannot be used with another version of
> > python (i.e.
> > python2.6, or python3.0, etc.). When you do try to import a .pyc
> > file from
> > a different version, it raises the exception you see (ImportError:
> > Bad magic
> > number). Try to find the default system python that compiled this
> > version
> > and use that (unless it's python < 2.6). The upcoming release will
> > not be
> > so finicky with regards to python version, and is a lot easier to
> > install
> > and run.
> >
> > Hope this helps,
> > Jason
> >
> >
> > > The script finds the inputparse.pyc but it seemed to be broken
> > because
> > > when I rename the inputparse.pyc I get
> > >
> > > Traceback (most recent call last):
> > > File "/opt/amber11/bin/MMPBSA.py", line 44, in <module>
> > > import inputparse, utils, alamdcrd # (2)
> > > ImportError: No module named inputparse
> > >
> > > So the MMPBSA.py might not work with Python3.0?Any thoughts on
> > this?
> > >
> > > Thx in advance
> > >
> > > Bernhard Poll
> > >
> > >
> > > Quoting Jason Swails <jason.swails.gmail.com>:
> > >
> > > > Perhaps try using MMPBSA.py (
> > > > http://ambermd.org/tutorials/advanced/tutorial3/py_script)
> > instead.
> > > > It may
> > > > print out a more helpful error message if there is, indeed,
> > one.
> > > >
> > > > Good luck,
> > > > Jason
> > > >
> > > > On Thu, Feb 17, 2011 at 11:38 AM, <poll.chemie.uni-hamburg.de>
> > > > wrote:
> > > >
> > > > > When I run this by hand no error occures and nothing strange
> > in
> > > > the
> > > > > output-file. I tried to run the MMPBSA without this snapshot
> > > > which
> > > > > didn't work as well.
> > > > >
> > > > > Quoting steinbrt.rci.rutgers.edu:
> > > > >
> > > > > > Hi,
> > > > > >
> > > > > > > /opt/amber11/bin/sander -O -i pbsa_com.in -o
> > > > > > pbsa_com.1.out -c
> > > > > > > ./snapshot_com.crd.1 -p ./1q3w_out.prmtop not successfu
> > > > > >
> > > > > > this is the actual command that failed. You could run it
> > by
> > > > hand
> > > > > > and see
> > > > > > what the output says. pbsa can sometimes crash on unusual,
> > > > large
> > > > > > (or even
> > > > > > random?) conformations, often the output file will contain
> > an
> > > > > > explanation
> > > > > > or a suggestion of what parameter to change.
> > > > > >
> > > > > > In the worst case, you could do the MMPBSA analysis
> > without
> > > > data
> > > > > > from that
> > > > > > one problematic snapshot.
> > > > > >
> > > > > > Thomas
> > > > > >
> > > > > > Dr. Thomas Steinbrecher
> > > > > > formerly at the
> > > > > > BioMaps Institute
> > > > > > Rutgers University
> > > > > > 610 Taylor Rd.
> > > > > > Piscataway, NJ 08854
> > > > > >
> > > > > > _______________________________________________
> > > > > > AMBER mailing list
> > > > > > AMBER.ambermd.org
> > > > > > http://lists.ambermd.org/mailman/listinfo/amber
> > > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > _______________________________________________
> > > > > 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
> > > >
> > >
> > >
> > >
> > >
> > > _______________________________________________
> > > 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
> >
>
>
>
>
> _______________________________________________
> 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 Tue Feb 22 2011 - 09:00:06 PST
