Hi,

We (Pierre-Yves Savard and I) have found a problem with the 
structure.create_diff_tensor_pdb() function.

If you use either results.read() or diffusion_tensor.init() to load a diffusion 
tensor and, then, display it, it is fine, except for the ellipsoid where some 
weird things happen...

Indeed, with the ellipsoid, you get a weird shaped tensor, depending on the 
angles input. With angles set to 0, the tensor is fine. With alpha and beta set 
to non-zero values, the shape of the tensor changes drastically as it becomes 
completely flat. With gamma set to a non-zero value, the shape is fine, with 
the tensor just turned around the Dz axis...

I submitted a bug report and attached 7 files:

tensor.pdb
    With all angles set to 0.

tensor__angles__alpha_90.pdb
    With alpha = 90 degrees.

tensor__angles__beta_90.pdb
    With beta = 90 degrees.

tensor__angles__gamma_90.pdb
    With gamma = 90 degrees.

tensor__angles__all_90.pdb
    With all angles set to 90 degrees.

tensor__angles__all_45.pdb
    With all angles set to 45 degrees.

tensor__angles__45_0_45.pdb
    With all angles set to 45, 0 and 45 (alpha, beta and gamma, respectively).


As you will see, the tensor is fine if both alpha and beta are equal (files:  
tensor__angles__all_45.pdb and tensor__angles__all_90.pdb), but weird shaped 
when alpha and beta are not equal (tensor__angles__45_0_45.pdb)...

These observations were made with the subversion repository. I verified with 
relax-1.3.1 / relax-1.3.2 / relax-1.3.3 and the same problem arises... so the 
problem as been there for a long time...

Shouldn't the tensor always keep a symmetric shape, with angles only changing 
the orientation, not the shape..?

Thanks for help.


Séb   :) 
-- Sébastien Morin PhD Student S. Gagné NMR Laboratory Université Laval
& PROTEO Québec, Canada



_______________________________________________
relax (http://nmr-relax.com)

This is the relax-users mailing list
relax-users@gna.org

To unsubscribe from this list, get a password
reminder, or change your subscription options,
visit the list information page at
https://mail.gna.org/listinfo/relax-users

Reply via email to