TI1     3   0.00000   0.00000   0.00000   0.25000   0.05811
TI2     3   0.14428   0.50000   0.52050   0.50000   0.03427
O3      2   0.09437   0.00000   0.25602   0.50000   0.05397
O4      2   0.22759   0.00000   0.68766   0.50000   0.13380
O5      2   0.08152   0.50000   0.89990   0.50000   0.00001

CELL   1.54184   14.09600  2.95200  4.89170  90.0000  92.1700  90.0000
ZERR   2          0.00200  0.00200  0.00100   0.0000   0.0500   0.0000
LATT  7
SYMM -X,Y,-Z

To transform the Cell, go into Model Transform to bring up the following options box.

Apply the transformation you wish to make. In this case abc-> cab.
You can press the TEST to check that the transformation of the cell is the one intended (c is the unique axis) before continuing.

Then select the space group you want to transform to, in this case A 1 2 2/m

Then press OK to continue leading to the following output and a new.ins and new.hkl with transformed, cell, co-ordinates and HKL data.

The resulting A 1 1 2/m based shelx INS file has the following transformed co-ordinates, cell and symmetry operators (and the the A 1 1 2/m based HKL file)

Ti1     3   0.00000   0.00000   0.00000   0.25000   0.05811
Ti2     3   0.52050   0.14428   0.50000   0.50000   0.03427
O3      2   0.25602   0.09437   0.00000   0.50000   0.05397
O4      2   0.68766   0.22759   0.00000   0.50000   0.13380
O5      2   0.89990   0.08152   0.50000   0.50000   0.00001

CELL   1.54184  4.8917 14.0960  2.9520   90.000   90.000   92.170
ZERR   2        0.0010  0.0020  0.0020    0.000    0.000    0.050
LATT  5
SYMM     -X ,    -Y ,     Z

Note from Louis Farrugia: "Are Transformations really that simple with WinGX?"

From: louis@chem.gla.ac.uk
To: Lachlan Cranswick l.m.d.cranswick@dl.ac.uk
Subject: Re: Some tutorials to vet(?) - Transformation??
Date: Thu, 24 Sep 1998 13:26:45 +0100


>One thing I am wondering if you could check a tutorial on Cell 
>transformation I have done - but not announced or linked yet at:
>  http://www.ccp14.ac.uk/tutorial/wingx/tranform/
>
>Is it really that simple with WinGX? - it is very unnerving as I am used to 
>transformation (for me) being a very nasty and painful experience.  And most 
>often with clueless arithmetic and matrix mistakes in it.

Yes it is that easy, in fact in my current version it is even easier

Say you had collected a structure in the non-standard orthorhombic setting
C2cb and wish to transform it to the standard setting. Using the menu 
option SPGR INFO (which incidentally is new and uses the GETSPEC algorithm)  
you type in C2cb  (spaces in string are no longer necessary!)
and you will see that C2cb is space group 41 with setting -cba.
Then got to transform menu and click on the -cba button (this dialog box has
also changed slightly to reflect the common matrices required for 
orthorhombic settings - sorry !) and select the standard setting of space 
group 41 Aba2. Then click OK - thats it!!!! The HKL file (possibly with 
direction cosines), the INS file CELL ZERR, LATT and SYMM and any
OMIT instructions with HKL's on them plus all atoms and adp's are correctly
transformed. The orientation matrix file name .OMX is also transformed, 
so that (most) absorption corrections may be carried out in the new setting.

Incidentally it is also now possible to enter alternative settings using the
conventions listed in file sginfo.dat (in files directory) so that the space 
group R-3 can be indicated as R-3:h for hexagonal axes or R-3:r for rhombohedral
axes. SPGR INFO wil give the correct LATT and SYMM cards now for both
options - remember when you are starting a structure using Initialise Files
to give the space group with its setting