December 2006 - cctbxbb - phenix-online.org

Re: [cctbxbb] Fail to extend space-group with centring translations
by Ralf W. Grosse-Kunstleve 11 Dec '06

11 Dec '06

Hi Luc, > If I understand correctly, space_group_type lets the user use any > space-group in a non-standard setting defined by a special change of > basis. space_group_type determines the change-of-basis matrix from a given setting to the reference setting. > Now your example would require the matrix of change of basis > to be rational whereas in the current implementation it can only be > integral. The change_of_basis_op class has both rational rotation and rational translation parts, with a common r_den for the 9 rotation matrix elements, and a common t_den for the 3 translation vector elements. The "r_den" is fixed at 1 only for the space_group class. Since space_group_type operates only on space_group objects, the limitation applies here too, but the resulting change-of-basis rotation parts certainly don't have to be integral (that would be a terrible limitation). > At least as far as I understand, from the code and your > comments. Thus, the only place where rational numbers would be > necessary is in those matrices of change of basis, isn't it? Now, > would it be possible to define a new class > space_group_irrational_setting which would provide services very > similar to space_group_type but accepting non-integral matrices for > the change of basis? You are correct. If you need to handle space groups with rational rotation matrices, the way forward is to implement a new class similar to space_group (I'd go for space_group_rational) with a little bit of cut-and-paste to replicate the group multiplication. However, I don't think it is worth redoing the entire space_group_type implementation to accommodate rational rotation parts. In fact, I think it would be a bad solution because of the runtime impact. Consider that any space group when transformed to the primitive setting will have only integral rotation matrices. I.e. all you need is an algorithm that figures out an "initial" change-of-basis matrix to a (any) primitive setting. Construct a standard space_group object with the transformed symmetry operations and feed it into space_group_type. Multiply the "initial" change-of-basis with the output of space_group_type, done. This way you get the desired generality with a moderate effort, without impacting the performance of the many established algorithms based on space_group. Cheers, Ralf ____________________________________________________________________________________ Yahoo! Music Unlimited Access over 1 million songs. http://music.yahoo.com/unlimited

2 1

Re: [cctbxbb] Fail to extend space-group with centring translations
by Ralf W. Grosse-Kunstleve 11 Dec '06

11 Dec '06

> Luc, This is a key item in Ralphs reply: > > "You cannot change the rotation denominator (fixed at 1) for the > matrices processed by sgtbx.space_group. Therefore you cannot work > with, e.g. the pseudo-orthorhombic C-centered setting of a hexagonal > space group. However, I never came across a situation where this > was interesting." > > If the design of the toolbox is too restrictive to permit the user to > invent any closed set of operators they want - well thats the end of it. > As we say in the UK, C'est la vie. Could you please outline a plausible example where the current implementation is a serious limitation? > There were lots of things we would have liked to add to CRYSTALS, but > there was no room in Bob Carruthers database, and extending it would > have been a major hasle. I am sure the cctbx is more flexible. > As you pointed out, there are work-arounds that can be applied by a > clever crystallographer, and the less clever ones won't even try to do > anything clever. It's a self-limiting situation! Much of what crystal > enginners want could be done if Oleg lets the user create an extended > atom parameter list, and put this into a new unit cell in P1. You can create any P1 cell you want, it is totally trivial: first use e.g. xray_structure.expand_to_p1(), then loop over the translations you want to duplicate the atoms. We can easily add methods to do it in one call. Note that the cctbx is open source, with the possibility of write access to the main source repository. Cheers, Ralf ____________________________________________________________________________________ Do you Yahoo!? Everyone is raving about the all-new Yahoo! Mail beta. http://new.mail.yahoo.com

2 1

Re: [cctbxbb] Fail to extend space-group with centring translations
by Ralf W. Grosse-Kunstleve 10 Dec '06

10 Dec '06

Hi Luc, You can make both examples work like this: from cctbx.sgtbx import * info = space_group_info("P2/a") g = info.group() g.expand_ltr(tr_vec((1,1,1),2).new_denominator(sg_t_den)) from cctbx import sgtbx s = sgtbx.space_group_symbols("P21/c") g = sgtbx.space_group(s) g.expand_ltr(sgtbx.tr_vec((1,0,1),3).new_denominator(sg_t_den)) g.expand_ltr(sgtbx.tr_vec((2,0,2),3).new_denominator(sg_t_den)) Any setting of any crystallographic space group can be processed via sgtbx.space_group, as long as the rotation matrix elements are integers. You may just have to be careful in choosing appropriate "t_den" (translation denominators). All matrices in a given space_group instance have to have the same t_den. I know it can be a hassle, but I don't consider it a practical limitation. Note that sg_t_den is just the default (12). You can set other values if necessary. You cannot change the rotation denominator (fixed at 1) for the matrices processed by sgtbx.space_group. Therefore you cannot work with, e.g. the pseudo-orthorhombic C-centered setting of a hexagonal space group. However, I never came across a situation where this was interesting. It would be possible to rewrite the sgtbx based on boost::rational<int>, but it is a big project. The advantage would be more flexibility, the disadvantage a performance decrease. If there is a strong reason I would go for more flexibility, but so far I've never found this worthwhile. BTW: A more convenient way of adding symmetry operations is: g.expand_smx("x+1/2,y+1/2,z+1/2") Cheers, Ralf ____________________________________________________________________________________ Do you Yahoo!? Everyone is raving about the all-new Yahoo! Mail beta. http://new.mail.yahoo.com

3 3

Fail to extend space-group with centring translations
by Luc Bourhis 09 Dec '06

09 Dec '06

Hello there, with the latest cctbx release, the following Python snippet fails from cctbx.sgtbx import * info = space_group_info("P2/a") g = info.group() g.expand_ltr(tr_vec((1,1,1),2)) The exception indicates that it is not possible to add a centring translation with a denominator of 2 to a group which does not have any centring translations with that denominator. Of course, in that case, that description is trivial because P2/a does not have any centring translations! My real intent was actually to introduce 1/3 centring translations so as to deal with a tripled cell. I am just citing the example above to show the extent of the problem. My original, failing, snippet was therefore from cctbx import sgtbx s = sgtbx.space_group_symbols("P21/c") g = sgtbx.space_group(s) g.expand_ltr(sgtbx.tr_vec((1,0,1),3)) g.expand_ltr(sgtbx.tr_vec((2,0,2),3)) It fails with the same class of exception. It seems like it is a deeply rooted limitation and that brings me to the bigger picture. I am currently working on an EPSRC grant to write the next generation of small molecule crystallographic software, a joint collaboration between Judith Howard in Durham and David Watkin in Oxford (all in UK, just in case…!). The CCTBX is obviously a seducing foundation for such a project. Thus we have tried to push it into corners which do actually matter to a significant minority of small molecule crystallographers. So I would like to know how difficult it would be to lift the limitation at hand. Would it trigger a cascade of changes throughout the sgtbx? Luc Bourhis

2 2

Obtaining symmetry information
by Konrad Hinsen 04 Dec '06

04 Dec '06

Dear All, I am trying to reconstruct an entire unit cell from an cctbx.xray.structure object that contains the scatterers in the asymmetric unit. Ideally, I'd call a method that does the work for me, but at least I would like to obtain a list of the symmetry transformations that must be applied. That information must be somewhere in there... but where? Konrad. -- --------------------------------------------------------------------- Konrad Hinsen Centre de Biophysique Moléculaire, CNRS Orléans Synchrotron Soleil - Division Expériences Saint Aubin - BP 48 91192 Gif sur Yvette Cedex, France Tel. +33-1 69 35 97 15 E-Mail: hinsen(a)cnrs-orleans.fr ---------------------------------------------------------------------

3 3