Re: [ccp4bb]: Reindex

[Marian Szebenyi <dms35@cornell.edu>]

***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***

It is true that there are several possible matrices to convert C2 into
P1. However, I don't think the one you give is one of them, unless I am
misreading it. Even if the matrix you use is not the one that gives the
most "standardized" P1 cell, a valid matrix should give **A** valid P1
cell, and the resolution should not change. I suggest applying the
following matrix in reindex:

1/2 -1/2 0
1/2 1/2  0
0    0   1,

i.e. h(P1) = 0.5*h(C2)-0.5*k(C2), k(P1) = 0.5*h(C2)+0.5*k(C2), l(P1) =
l(C2)

Marian Szebenyi 

"Eleanor J. Dodson" wrote:
> 
> ***  For details on how to be removed from this list visit the  ***
> ***          CCP4 home page http://www.ccp4.ac.uk         ***
> 
> Miha Andrejasic wrote:
> > ***  For details on how to be removed from this list visit the  ***
> > ***          CCP4 home page http://www.ccp4.ac.uk         ***
> >
> > Hello everybody!
> >
> > I posted question regarding reindex few days ago. Now I have similar but
> > very specific question.
> >
> > I am looking for transofmation matrix (3 by 3) for converting hkl
> > indexes between various specegroups.
> > I found matrix for conversion between P1 into C2. I want to convert C2
> > into P1.
> >
> > P1 -> C2: 1/2 1/2 -1/2  -1/2 1/2 1/2  1 0 0
> >
> > I inverted the matrix to get reverse:
> >
> > C2 -> P1: 0 0 1  1 1 0  -1 1 1
> >
> > But with no succes. When I apply this matrix on reflexes (after
> > conversion of C2 into P1 with SFTOOLS) I get resolution in P1 0.5A.
> > Without matrix was 0.8A. Real resolution in C2 is 2.5A.
> >
> > Any help will be mostly apreciated.
> >
> > Thank you
> 
>   There isnt just one matrix - it will depend on your cell dimensions..
> 
>   It is never easy - I start by looking at hklview section hk0 h0l 0kl
> etc, and seeinf if I can spot extra symmetry..
> 
> Then run axissearch and see if there are other condintions of a*, b* and
> c* which have suitable angles, etc..
> 
>   Then you can use reindex to get your matrix.
> 
> Eleanor