[MMTK] Re: MMTK Windows Problems

Robin Maytum R.M.Maytum at kent.ac.uk
Sat Jul 12 01:48:59 EDT 2003

Hi !

I'm also just in the pangs to tying to MMTK to work on Windows so I'm no
expert, but have come across a few of the same problems. 

> I had valued the PDBVIEWER in order to have eg
> >>>os.environ['PDBVIEWER']
> 'C:\\Program Files\\University of Illinois\\VMD\\vmd'

MMTK under Windows does not appear to use the PDBVIEWER variable.
Instead it uses the Windows API to call whichever program is associated
with PDB files.

If you look under Explorer and check the file association of your PDB
files - link this with your viewer program and you should find that the
visualisation program is called correctly.

> I am trying this example Ffor Visulaization in MMTK-examples modified for
> VMD instead of VRML:
> _________________________________________________
> from Scientific.Visualization import VMD; module = VMD
> protein = Protein('insulin')
> center, inertia = protein.centerAndMomentOfInertia()
> mass = protein.mass()
> diagonal, directions = LinearAlgebra.eigenvectors(inertia.array)
> diagonal = Numeric.sqrt(diagonal/mass)
> graphics = protein.graphicsObjects(graphics_module = module,
>                                    model = 'backbone', color = 'red')
> valines = protein.residuesOfType('val')
> sidechains = valines.map(lambda r: r.sidechain)
> graphics = graphics + sidechains.graphicsObjects(graphics_module = module,
>                                                  model='wireframe',
>                                                  color = 'blue')
> for length, axis in map(None, diagonal, directions):
>   graphics.append(module.Arrow(center, center+length*Vector(axis), 0.02,
>                                  material=module.EmissiveMaterial('green')))
> scene = VMD.Scene(graphics)
> scene.view()
> >>>
> The system cannot find the path specified.
> >>>
> ______________________________________________

> Why? Is the path just the executable of VMD with the directory specified
> please?

This is where the Windows API is trying to call whichever program is
associated with PDB files - if not program is associated it comes up
with this error.
> I'd like to ask also if there is any answer to why the MolecularDynamics
> give MEMORYERROR in Windows installations:

Sorry can't help you with this. At the moment my installation crashes
out when I try and load the minimizer module

>>>From MMTK.Minimization import ConjugateGradientMinimizer

Causes the program to crash with a Windows exception error

It seems to be specific to this part of the program as, for example, you
can import a water into an Amber universe and print the energy terms

I can also get it to visualize the water by calling up Rasmol which I
have associated with PDB files :-)

Any hints of getting the minimizer to work would be gratefully received.

As a background I'm running

MMTK 2.2
Python 2.2.3
Numeric Python 23.0 (I've tried recompiling 22.0 with no difference)
Scientific Python 2.4.3
NetCDF 3.5.0

Compiled using MinGW

I've been trying to put together a set of 'from scratch' instructions to
get MMTK working on a Windows system which I'm happy to forward once
I've got the whole thing working properly - I've found you often try so
many different things to get the system to work during an instalation
that it's often difficult to trace back everything you have done to work
out what has actually solved the problem !

I have to say I'm looking towards installing a Linux system in the near
future to try and get the whole thing working :-/

Thanks for any help !


Dr Robin Maytum, Dept Biosciences, University of Kent, Canterbury, UK

More information about the mmtk mailing list