A nice detailed analysis! This is exactly what's happening. I'm confused
why this particular issue exists, since I distinctly remember addressing
this, but I must not have committed it :(. While this patch works, I'm a
little concerned about what may happen in parallel. To avoid having to
know when each thread knows about everything, I typically quit after a
fatal error by throwing an uncaught exception (I modified the excepthook to
optionally suppress the traceback and call MPI.Abort to kill all threads).
I'll create a patch for this.
Back to the OP, though -- all this patch will do is provide a more helpful
error message -- the problem is that cpptraj is not installed, and it needs
to be for MMPBSA.py to work. Make sure you install all of AmberTools 12,
and that each of the mmpbsa_py tests pass, before trying to perform your
own simulations.
As a side note, if cpptraj _is_ installed, make sure AMBERHOME is properly
set (to your AmberTools installation, NOT your older Amber installation --
see
http://jswails.wikidot.com/installing-ambertools-12-and-older-amber for
details).
HTH,
Jason
On Tue, Apr 17, 2012 at 8:28 AM, Jan-Philip Gehrcke <jgehrcke.googlemail.com
> wrote:
> Huhu,
>
> "OSError: [Errno 2] No such file or directory" in context of subprocess
> means that the executable path provided to `subprocess.Popen()` is not a
> valid path. This is contradictory to the message "cpptraj found! Using
> cpptraj". Likewise, there must be a problem in the source code.
>
> I went through AmberTools12's MMPBSA source, and if I am not overseeing
> something elementary, I found this problem. However, I did not compile
> AmberTools12 myself so far and did not run any tests. So in case I've
> overseen something, I may have identified the problem wrongly which also
> renders the proposed solution to nonsense :-)
>
>
> Let's look at line 37 of findprogs.py (in function `find_progs()`):
>
> print '%s found! Using %s' % (prog, str(my_progs[prog]))
>
> which in Kumar's case resulted in "cpptraj found! Using cpptraj".
>
>
> From the code in context, `prog` is the 'required' program name, in this
> case "cpptraj". "cpptraj found" actually means cpptraj is
> required/requested. It does not mean that the executable has been found so
> far.
>
> `str(my_progs[prog])` prints what the `__str__` method of class
> `ExternProg` returns:
>
>
> def __str__(self):
> if self.full_path: return self.full_path
> else: return self.prog_name
>
> `self.full_path` is what `which(self.prog_name, self.search_path)`
> returns. This function returns `None` if the requested executable was
> neither found in a) $AMBERHOME/bin nor in $PATH. Hence, if the program is
> not found by `which()`, `self.full_path` is `None` and `self.prog_name` is
> printed. Which does not make sense.
>
> This means that "cpptraj found! Using cpptraj" is printed even if the
> cpptraj executable is not found. The same is true for any other required
> program.
>
> A `None` returned by `which()` indicates that the executable was not
> found. However, I cannot find any place where a `None` returned by
> `which()` results in an error message. I think that `find_progs()` should
> return `False` or `None` in case one of the requested executables could not
> be found. Then, in MMPBSA.py(cc), after line 497:
>
> external_progs = find_progs(INPUT)
>
> we should check for `external_progs` being `None` or `False`. I attached
> two patch files implementing these suggestions. Again, I did not test
> anything of this.
>
> This is only a suggestion and maybe it helps,
>
> Jan-Philip
>
>
>
>
>
>
> On 04/17/2012 12:51 PM, Kumar, Sonu wrote:
>
>> Hi Jason,
>>
>> I tried with 'debug_printlevel=2' and now it is giving this error:
>>
>> Reading command-line arguments and input files...
>> Loading and checking parameter files for compatibility...
>> mmpbsa_py_energy found! Using mmpbsa_py_energy
>> cpptraj found! Using cpptraj
>> Preparing trajectories for simulation...
>> File "/usr/local/AmberTools/bin/**MMPBSA.py", line 516, in<module>
>> str(external_progs['cpptraj'])**)
>> File "/usr/local/AmberTools/bin/**MMPBSA_mods/make_trajs.py", line 57,
>> in make_trajectories
>> traj = Trajectory(FILES.solvated_**prmtop, FILES.mdcrd, cpptraj)
>> File "/usr/local/AmberTools/bin/**MMPBSA_mods/make_trajs.py", line
>> 453, in __init__
>> self.Query()
>> File "/usr/local/AmberTools/bin/**MMPBSA_mods/make_trajs.py", line
>> 577, in Query
>> process = Popen([self.exe, self.prmtop], stdin=PIPE, stdout=PIPE)
>> File "/usr/lib64/python2.6/**subprocess.py", line 595, in __init__
>> errread, errwrite)
>> File "/usr/lib64/python2.6/**subprocess.py", line 1092, in
>> _execute_child
>> raise child_exception
>> OSError: [Errno 2] No such file or directory
>> Exiting. All files have been retained.
>>
>> Thanks,
>> Sonu
>>
>>
>> On Apr 12, 2012, at 2:56 PM, Jason Swails wrote:
>>
>> Can you add "debug_printlevel=2" to the&general section of your MMPBSA
>>>
>>> input file and retry it? That will provide a full traceback so I can see
>>> where the error is really occurring.
>>>
>>> Thanks!
>>> Jason
>>>
>>> On Wed, Apr 11, 2012 at 5:24 PM, Kumar, Sonu
>>> <sonu.kumar.dkfz-heidelberg.de**>wrote:
>>>
>>> Hi All,
>>>>
>>>> I am using MMPBSA.py module of AmberTool12 which is on my institute
>>>> cluster for energy calculation, but getting an error?
>>>>
>>>> Here is my input file:
>>>> mmpbsa.in
>>>> Input file for running PB and GB in serial
>>>> &general
>>>> endframe=50, keep_files=2,
>>>> /
>>>> &gb
>>>> igb=2, saltcon=0.100,
>>>> /
>>>> &pb
>>>> istrng=0.100,
>>>> /
>>>>
>>>> /usr/local/AmberTools/bin/**MMPBSA.py -O -i mmpbsa.in -o
>>>> FINAL_RESULTS_MMPBSA.dat -sp complex_solvated.prmtop -cp complex.prmtop
>>>> -rp
>>>> rec.prmtop -lp lig.prmtop -y *.crdbox
>>>>
>>>> Error which I am getting after running above command:
>>>>
>>>> Reading command-line arguments and input files...
>>>> Loading and checking parameter files for compatibility...
>>>> mmpbsa_py_energy found! Using mmpbsa_py_energy
>>>> cpptraj found! Using cpptraj
>>>> Preparing trajectories for simulation...
>>>> OSError: [Errno 2] No such file or directory
>>>> Exiting. All files have been retained.
>>>>
>>>>
>>>> It would be great if someone help me in this..
>>>>
>>>> Thanks
>>>> Sonu
>>>>
>>>> ----------------
>>>> Sonu Kumar, M.Sc
>>>>
>>>>
>>>>
>>>> ______________________________**_________________
>>>> AMBER mailing list
>>>> AMBER.ambermd.org
>>>> http://lists.ambermd.org/**mailman/listinfo/amber<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<http://lists.ambermd.org/mailman/listinfo/amber>
>>>
>>
>> ----------------
>> Sonu Kumar, M.Sc
>> GlycoImmunology Lab(D015)
>> German Cancer Research Center(DKFZ),
>> Heidelberg, Germany
>>
>> sonu.kumar.dkfz.de
>>
>>
>> ______________________________**_________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/**mailman/listinfo/amber<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 Apr 17 2012 - 07:30:03 PDT
