Why is this Blog called P212121?
Mar 22, 2009
Why is this blog called P212121?
P212121 is the most common space group in macromolecular crystallography. I have not yet found an explanation of why P212121 has resulted in more deposits than any other space group in the PDB. As you may already know there are a total of 65 possible space groups for macromolecular crystallography. P212121 comprises almost 25 percent of those deposited within the PDB. This strange occurrence inspired the naming of this blog.
Stephen Curry
March 23rd, 2009 at 9:53 AM #
I haven’t counted recently but I thought there were 65 spacegroups available to proteins. Has one of them stormed out?
admin
March 23rd, 2009 at 11:58 AM #
Thanks Stephen for catching that error. I update the post from 64 to 65, I wonder if that the reason I can’t solve some data sets
P1
April 21st, 2009 at 10:07 PM #
Look at Todd Yeates , See Wukovitz and Yeates. 1995. Why Protein Crystals Favour Some Space Groups Over Others. Nature Structural Biology 2, 1062-1067.
admin
April 22nd, 2009 at 8:26 AM #
Thanks for the reply! I have been getting a lot more emails than comments and need to do a better of job of posting them.
My favorite explanation so far addressing this question is by ‘Dave’.
—–
Isn’t P212121 the most common space group precisely because it has all 21 screw axes?
Kitaigorodskii set out how screw axes allow better packing, due to e.g. paired charges, knob-into-holes, etc.
See Kitaigorodskii, A. I. Organic Chemical Crystallography; Consultant’s
Bureau: New York, 1961. (English translation of the Russian original
published by Press of the Academy of Sciences of the USSR, Moscow,
1955).
Also http://pubs.acs.org/doi/pdf/10.1021/cm00044a010, Towards a Grammar of Crystal Packing, Carolyn Pratt Brock, and Jack D. Dunitz, Chem. Mater., 1994, 6 (8), 1118-1127* DOI: 10.1021/cm00044a010, which references the above & more
——
Todd
July 8th, 2009 at 8:34 PM #
The issue of space group frequencies for macromolecules (which is not the same as for small molecules, which is mainly the context in which it was discussed by Kitaigorodskii and Brock & Dunitz), and the answer to the P212121 question, is covered at some length at:
http://www.doe-mbi.ucla.edu/~yeates/old_space_group_freq.html
The original work is Wukovitz and Yeates (1995). Nat. Struct. Biol. 2, 1062-7.
A recent Commentary by Brian Matthews emphasizing the successful predictions of that work can be found in Matthews (2009). Protein Sci. 18, 1135-8.
Sean
July 8th, 2009 at 8:53 PM #
Thanks for the references! I have looked at Kitaigorodskii a couple of months ago and from memory thought he referred to non-convalent interactions as ‘bumps and bulges’ (as well as crystal packing that you mention on your site).