Hi, If you go to
http://cci.lbl.gov/cctbx/explore_symmetry.html fill out any space group (F432), and search for: Additional generators of Euclidean normalizer to get Number of structure-seminvariant vectors and moduli: 1 Vector Modulus (1, 0, 0) 2 The seminvariants for F432 are thus (1/2, 0, 0) so the only unique origin shift to consider here is (1/2,0,0). The 1/4 solution do not arise in F432 as can be seen in http://scripts.iucr.org/cgi-bin/paper?a10657 (table 4) as well. Giacovazzo actually lists 1/2 1/2 1/2, which is symmetry equivalent to 1/2 0 0 (F centering). In the cctbx, these seminvariants are derived from first principles and do work as well when you have things in weird settings. One nice application of doing this from first principles over using lookup tables, is (for instance) finding "pseudo" origin shifts in the presence translational NCS. For instance the origin shifts in P21: ------------------ Number of structure-seminvariant vectors and moduli: 3 Vector Modulus (1, 0, 0) 2 (0, 1, 0) 0 <- continuous shift along y (0, 0, 1) 2 Inversion through a centre at: 0,0,0 ------------------ and in P21 with added (pseudo) (x+1/2,y,z), one gets ------------------ Number of structure-seminvariant vectors and moduli: 3 Vector Modulus (1, 0, 0) 4 (0, 1, 0) 0 (0, 0, 1) 2 Inversion through a centre at: 0,0,0 ------------------ This indicates that you have an additional origin shift of (1/4,0,0) when you have (x+1/2,y,z) as part of the symmetry. If (x+1/2,y,z) is a 'pseudo' operator, then the origin shift of (1/4,0,0) will be pseudo as well. This can lead to interesting molecular replacement problems (Andrey Lebedev has done some nice work there). Cheers, Peter 2010/1/27 James Holton <jmhol...@lbl.gov>: > Francois Berenger wrote: >> >> It correspond to what is found at the end of James Holton's >> origins.com script >> http://bl831.als.lbl.gov/~jamesh/pickup/origins.com >> so I guess it should be correct. > > Uhhh... >> >> He also says how he found them: >> <origins.com snippet> >> # TABLE OF ALLOWED ORIGIN SHIFTS >> These origin shifts were determined emprirically using 100 randomly-placed >> atoms >> that were shifted around with pdbset and checked with SFALL for identical >> amplitudes to the 0 0 0 origin. They should be correct for the CCP4 >> convention >> of symmetry. (I.E. R3 and R32 have hexagonal indexing) >> </origins.com snippet> > > Okay, I just went back to my notes on this. I should admit that in my > completely brain-dead allowed-origin-shift search described above I > originally found no "allowed" shifts for F432 (not sure why, but maybe > because it was last on my list). Years later, I think it was Peter Zwart who > pointed out to me that I was missing a few of what ought to be allowed > origins. As I recall, at least one of them passed my pdbset/sfall test, so > I just assumed I must have done something wrong and added everything in the > CCP4 document and ITC Vol B to the list. For purposes of the origins.com > script, I decided to err on the side of having it try things that may or may > not work. If it finds a match, then great! Right? > > Perhaps this was not wise of me. I just tried shifting the 2fka PDB file by > "SHIFT FRAC 0.25 0.25 0.25" with pdbset and comparing the resulting Fs from > sfall to those of the un-shifted PDB. They are ~30% different. It could be > that this is a bug in sfall or pdbset (doubtful), but I would say that > pragmatically, this is not an "allowed origin shift" for F432. However, > 0,0,0.5 and 0,0.5,0.5 and all the other half-cell combinations do work! > Guess I missed those! I have now updated my origins.com script. It > appears that the allowed origin shifts for F432 and F23 are not the same, as > I had previously thought. > >> >>> plus of course the symmetry-equivalent origins generated from these 4 by >>> the space-group centring (F) translations: >>> >>> 0.0000 0.5000 0.5000 >>> 0.5000 0.0000 0.5000 >>> 0.5000 0.5000 0.0000 >> >> I can see these in syminfo.lib. >> >> However, from what you say, I understand that only 2x3 possible >> origins with each coordinate being 0 or .5 should be accepted by >> reforigin. >> But in my test it accepted all 8 possible combinations of 0 and .5 >> that I artificially introduced in my translated test PDBs: >> >> m...@myps:2fka# grep Frac run.log | sort | uniq >> Fractional origin shift: 0.0000 0.0000 0.0000 >> Fractional origin shift: 0.0000 0.0000 0.5000 >> Fractional origin shift: 0.0000 0.5000 0.0000 >> Fractional origin shift: 0.0000 0.5000 0.5000 >> Fractional origin shift: 0.5000 0.0000 0.0000 >> Fractional origin shift: 0.5000 0.0000 0.5000 >> Fractional origin shift: 0.5000 0.5000 0.0000 >> Fractional origin shift: 0.5000 0.5000 0.5000 >> >> Should I be worried? > > At this point, I would say no. Reforigin seems to be doing it right. > > I think some of the confusion might be arising because an "alternate origin" > and a fractional coordinate shift that gives you the same structure factor > amplitudes may or may not be the same thing. Some shifts change the phase, > but not the amplitude so whether or not they are "alternate" depends on what > you are trying to do. > -James Holton > MAD Scientist > >> >> Thanks, >> Francois. >> >>> So there will be 12 in all, which I think include the ones you >>> mentioned. >>> >>> If you're going by >>> http://www.ccp4.ac.uk/dist/html/alternate_origins.html then you should >>> be aware of a very recent BB discussion in which it was pointed out that >>> the entries for F222, F23, F432 and possibly others are incomplete. >>> >>> Eleanor has given me the task of checking & correcting this particular >>> documentation, until then don't trust it! >>> >>> Of course you shouldn't trust reforigin either, just as you shouldn't >>> trust any program until you have verified that the results are sensible, >>> but I think in this particular the fault doesn't lie with reforigin. >>> >>> Cheers >>> >>> -- Ian >>> >>>> -----Original Message----- >>>> From: owner-ccp...@jiscmail.ac.uk [mailto:owner-ccp...@jiscmail.ac.uk] >>> >>> On >>>> >>>> Behalf Of Francois Berenger >>>> Sent: 27 January 2010 06:50 >>>> To: CCP4BB@JISCMAIL.AC.UK >>>> Subject: reforigin on 2FKA >>>> >>>> Hello, >>>> >>>> I am playing with ccp4's reforigin to verify some MR solutions. >>>> >>>> If I translate a copy of the pdb.org's PDB 2FKA (from spacegroup F432) >>>> by +/-0.5 fractional in any unit cell direction, then reforigin will >>>> find back this translation and consider it as valid for this >>> >>> spacegroup. >>>> >>>> But for this spacegroup I should find only (0,0,0) or (1/2,1/2,1/2) >>>> as possible alternate origins. >>>> >>>> Does this mean that I can't trust reforigin and that I must filter >>>> out its results to retain only the valid ones? >>>> >>>> Thanks, >>>> Francois. >>> >>> >>> >>> Disclaimer >>> This communication is confidential and may contain privileged information >>> intended solely for the named addressee(s). It may not be used or disclosed >>> except for the purpose for which it has been sent. If you are not the >>> intended recipient you must not review, use, disclose, copy, distribute or >>> take any action in reliance upon it. If you have received this communication >>> in error, please notify Astex Therapeutics Ltd by emailing >>> i.tic...@astex-therapeutics.com and destroy all copies of the message and >>> any attached documents. Astex Therapeutics Ltd monitors, controls and >>> protects all its messaging traffic in compliance with its corporate email >>> policy. The Company accepts no liability or responsibility for any onward >>> transmission or use of emails and attachments having left the Astex >>> Therapeutics domain. Unless expressly stated, opinions in this message are >>> those of the individual sender and not of Astex Therapeutics Ltd. The >>> recipient should check this email and any attachments for the presence of >>> computer viruses. Astex Therapeutics Ltd accepts no liability for damage >>> caused by any virus transmitted by this email. E-mail is susceptible to data >>> corruption, interception, unauthorized amendment, and tampering, Astex >>> Therapeutics Ltd only send and receive e-mails on the basis that the Company >>> is not liable for any such alteration or any consequences thereof. >>> Astex Therapeutics Ltd., Registered in England at 436 Cambridge Science >>> Park, Cambridge CB4 0QA under number 3751674 > -- ----------------------------------------------------------------- P.H. Zwart Beamline Scientist Berkeley Center for Structural Biology Lawrence Berkeley National Laboratories 1 Cyclotron Road, Berkeley, CA-94703, USA Cell: 510 289 9246 BCSB: http://bcsb.als.lbl.gov PHENIX: http://www.phenix-online.org CCTBX: http://cctbx.sf.net -----------------------------------------------------------------
