Its may be an interesting question of pseudosymmetry. The best thing would be to find a local mentor who could have a good hands-on look, but here are some thoughts:
>From the space group tables you will see that P212121 has half as many >asymmetric units as C2221, and C2221 has crystallographic twofold axes that >your dimer could lie on. So both space groups are perfectly consistent with >the protein being a dimer. (You must be able to see the symmetric dimer when >you rebuild your model in C2221 during the refinement process?) The R-factor might be higher in P212121 because you're refining two molecules instead of one, and thus there are more degrees of freedom. If the space group does turn out to be P212121, you should probably find a local expert for detailed refinement advice. The centering in C2221 should give a different pattern of systematically missing spots, so you should have a carefull look at the original images and the output of the scaling in C2221, and see if the additional missing spots are really missing or not. (Don't forget the axes may be permuted between the two indexing systems). If they're present but weak, its P212121. In that case, you should note that the systematically weak spots will distort the standard twinning statistics. Hope that helps! Phoebe ===================================== Phoebe A. Rice Dept. of Biochemistry & Molecular Biology The University of Chicago phone 773 834 1723 http://bmb.bsd.uchicago.edu/Faculty_and_Research/01_Faculty/01_Faculty_Alphabetically.php?faculty_id=123 http://www.rsc.org/shop/books/2008/9780854042722.asp ---- Original message ---- >Date: Wed, 1 Dec 2010 16:17:58 +0800 >From: CCP4 bulletin board <CCP4BB@JISCMAIL.AC.UK> (on behalf of Xiaopeng Hu ><huxp...@mail.sysu.edu.cn>) >Subject: [ccp4bb] Space group and R/Rfree value >To: CCP4BB@JISCMAIL.AC.UK > >Dear all, > >I am working on a data-set (2.3A) and the space group problem bothers me a >lot.The space group of the data-set could be C2221 or P212121, since our >protein functions as a dimer, and P212121 gives two molecular in the >asym-uint, I think P212121 is more reasonable than C2221.However with C2221, I >can refine the R/Rfree to 20/24 or lower, while with P212121 only to 26/30. >Also Phenix points out that the crystal is probably a twin with P212121 but is >OK with C2221. > >I am not a real crystallographer, perhaps this problem is stupid, any help >will be appreciated!! > >Best wishes, > > >Xiaopeng Hu
