Re: [Open Babel] Translation of crystallographic coordinates intoCartesian

[Vincent Favre-N. <vi...@us...>]

	Hello again,

   OK, I spent some time looking at the formulas used and I believe
there are small mistakes in your Excel files:

  The most important is that the formula used for the y coordinate has a
wrongly placed parenthesis.

You wrote in the IT2 xls file for the y coordinate:
=D3*SIN(H3)*J3+(E3*(COS(F3)-COS(G3)*COS(H3)/SIN(H3)))*K3

it should be :

=D3*SIN(H3)*J3+(E3*(COS(F3)-COS(G3)*COS(H3))/SIN(H3))*K3

  Also, in the OB xls file, you wrote for the y coordinate:
=D4*SIN(H4)*J4-(E4*SIN(G4)*(COS(G4)*COS(H4)-COS(F4))/SIN(G4)*SIN(H4))*K4
it should be:
=D4*SIN(H4)*J4-(E4*SIN(G4)*(COS(G4)*COS(H4)-COS(F4))/SIN(G4)/SIN(H4))*K4


  Once this is corrected, the coordinates computed by the OB code are
identical to the IT2 matrix.

  Less important, the alpha beta gamma parameters must be given in
radians (but that does not affect the validity of the test) - unless
LibreOffice behaves differently than Excel in this matter, of course.



  As for the pdb file, I have regenerated it from your cif file using OB
2.3.0 and it seems to give the same coordinates as the IT2 matrix used
in OB - I do not know why you pdb file is different - did you use a
special option ?

  Attached: the modified IT2 and OB .xls files, and the pdb file
generated, as well as a python script for quick tests.

  Denis, could you double-check all the above ? It's a bit late here and
I had a long day/week...

      Vincent