[MMTK] Problem with Scientific python versions

[Dr. Jean-Didier Maréchal]

Hi Konrad,

I have something weird happening. 
On a computer that I let without work during a while, I installed the last
version of scientific python (2.9) and then wanted to install MMTK. However,
when I want to build MMTK I have the message:
MMTK needs ScientificPython 2.6 or higher... (something I thought I did).

I re-executed the installation of ScientificPython 2.9 and repeated the
installation of MMTK. Same problem.  And indeed, when asking for the version
of scientific python:
>From Scientific import __version__ as scientific_version
print scientific_version

the answer is .... 2.4.11

I guessed I made a mistake in reinstalling scientific python, but could you
give me a idea of where to beginning with to solve this problem?

All the best,
JD


-----Mensaje original-----
De: Konrad Hinsen [mailto:hinsen at cnrs-orleans.fr] 
Enviado el: lunes, 04 de mayo de 2009 8:27
Para: JeanDidier.Marechal at uab.es; mmtk forum
Asunto: Re: [MMTK] keeping the normal modes array in a file

On 01.05.2009, at 14:55, Jean Didier Pie Marechal wrote:

> I want to keep the normal modes vectors in a  file so that I can  
> use them in an outside program.
> I have to same requests:
>
> 1. what is the correct syntax to do so? I have for the moment in my  
> script an export loop like this:
> for n in [6, 7, 8]:
>   universe.writeToFile('mode_%d_plus.pdb' % n, conf+scale*modes[n])
>   universe.writeToFile('mode_%d_minus.pdb' % n, conf-scale*modes[n])
>   universe.writeToFile('file' %n, modes[n].array)

If you want PDB files, that's the way to go.

> 2. IN the only discussion in the archives regarding this type of  
> queries,  Konrad mentioned to be careful in using the vectors for  
> outside MMTK purposes. Could you tell me why?

Because if all you have is the arrays, you don't know which entry  
corresponds to which atom. The internal atom order is not deducible  
from the way the universe was constructed. However, if you export  
your displacement vectors to PDB files, you should not have any  
problem with that.

One caveat: PDB export is intended for configurations, not for  
displacement vectors. For an infinite universe (as you are probably  
using for normal modes), this makes no difference. But if you  
calculate normal modes in a periodic universe, the transformations  
applied during PDB export can ruin your displacement vectors. In that  
case, you should better export the displaced configurations, as you  
do in your first two lines.

Konrad.
--
---------------------------------------------------------------------
Konrad Hinsen
Centre de Biophysique Moléculaire, CNRS Orléans
Synchrotron Soleil - Division Expériences
Saint Aubin - BP 48
91192 Gif sur Yvette Cedex, France
Tel. +33-1 69 35 97 15
E-Mail: hinsen at cnrs-orleans.fr
---------------------------------------------------------------------