CNS holds the space group information in the file:
which on my system is:
and this does not contain P2(1)22(1) because at least some
crystallographers consider this to be a non-standard setting of
P2(1)2(1)2 . CNS reflects this, but PHENIX and large chunks of the CCP4
suite are rather more liberal in their policies. PHASER, in particular,
can report solutions for "non-standard" space groups while testing all
possibilities in primitive orthorhombic.
You might look at the CCP4bb bulletin board archives for a rather more
extensive discussion about what constitutes "standard" space groups in
primitive orthorhombic, although this is not going to change the default
behavior of CNS.
If you want to use CNS in this manner you'd need to reindex your data to
put the pure 2-fold axis on the c axis (h,k,l) -> (l,h,k), regenerate
a reflection file, change the order of your cell dimensions, and rotate
your model accordingly. If you kept your master reflection files in
CCP4/MTZ format the program REINDEX will handle most of this easily and
probably make the correct changes to cell dimensions and space group.
> Hello all,
> I am trying to make a composite-SA-omit map. My space group is P21221,
> and I specify that in my .inp file. However, when I try to run the job,
> it fails and gives me the error message, "space group "P2(1)22(1)" does
> not exist in the library." When I look at the spacegroup library that
> CNS calls, I see that the closest thing it has is P21212, but not mine.
> So what can I do about this? (I have built and refined using phenix, and
> this is my first time using CNS).