From pre@mrc-lmb.cam.ac.uk Wed Dec 7 12:02:45 2005 Date: Fri, 4 Nov 2005 11:22:13 +0000 From: Phil Evans To: ccp4@dl.ac.uk Subject: [ccp4]: forwarded message from Phil Evans [ Part 1: "forwarded message" ] Date: Fri, 4 Nov 2005 11:12:37 +0000 From: Phil Evans To: Tim Gruene Subject: Re: ISYM This is wrong ISYM is the symmetry number used to reduce the reflection to the asymmetric unit These are numbered from the spacegroup symmetry operators stored in the MTZ file, but modified so that odd numbers are +(hkl) ie using the operators themselves (assumed in this context to be non-centrosymmetric) ie - if the j'th operator (j=1,NsymP) was used, then ISYM = 2j-1 - if the j'th operator (j=1,NsymP) was used + inversion, then ISYM = 2j Phil Tim Gruene writes: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hello, > > my ccp4-documtation reads > (file:///usr/local/crystallography/ccp4-5.0.2/html/mtzformat.html#itemnames): > M/ISYM > Partial flag and symmetry number. 256M+Isym, where M is the partiality > flag (0 for full, 1 for partial) and the `symmetry number' Isym is > normally 0 (with F=(F+ - F-)/2 for anomalous data with both components > measured) but is 1 if F=F+ or 2 for F=F-. > > This is consistent with even ISYM for -(hkl) and odd for hkl. But as I > understand it, M/ISYM can only have 6 possible values: > 0, 1, 2, 256, 257, 258 > > But when I look at scala files, I see a much larger variety of numbers > (e.g. 0-24), whyfore I assumed scala was doing something very different > from what the ccp4-documentation suggests. With this range, how can one > tell the difference between F=(F+ - F-)/2 and F=F-, if another (unknown) > even number is added? > > Tim > > - -- > Tim Gruene > Institut fuer anorganische Chemie > Tammannstr. 4 > D-37077 Goettingen > > GPG Key ID = A46BEE1A > > > On Fri, 4 Nov 2005, Phil Evans wrote: > > > > > > > ISYM odd is hkl > > ISYM even is -h-k-l > > > > eg in P1 reduced reflections have ISYM = 1 (hkl) or 2 (-h-k-l) > > > > eg in the log file from Scala > > Spacegroup 80 "I 41" > > Original indices for reflection hkl with symmetry number ISYM > > > > Bijvoet positive > > ISYM ISYM ISYM ISYM > > ISYM 1 +h,+k,+l 3 -k,+h,+l 5 -h,-k,+l 7 +k,-h,+l > > > > Bijvoet negative > > ISYM ISYM ISYM ISYM > > ISYM 2 -h,-k,-l 4 +k,-h,-l 6 +h,+k,-l 8 -k,+h,-l > > > > > > Is this contrary to documentation? If so this needs to be corrected > > > > Phil > > > > George M. Sheldrick writes: > > > Dear Phil, Andrew and Harry, > > > > > > As you know I got the job of producing some trivial software for use > > > with the BioXhit test crystal project. This has been proving quite > > > useful, and some of the BioXhit users have asked us to add statistics on > > > the anomalous signal. This was no problem for input from XDS and HKL2000 > > > but we found it more difficult to extract the information from the mtz > > > file after SCALA that still contains unmerged data (which I need for the > > > statistics) because at this point, although there is a separate record > > > for each measured reflection, the indices have already been transformed > > > to the standard setting that is the same for equivalents of h,k,l and > > > for equivalents of -h,-k,-l. > > > > > > Am I right in thinking that, contrary to the CCP4 documentation, an odd > > > value of ISYM is used to refer to an equivalent of h,k,l and an even > > > value to refer to an equivalent of -h,-k,-l ? > > > > > > Best wishes, George > > > > > > -- > > > Prof. George M. Sheldrick FRS > > > Dept. Structural Chemistry, > > > University of Goettingen, > > > Tammannstr. 4, > > > D37077 Goettingen, Germany > > > Tel. +49-551-39-3021 or -3068 > > > Fax. +49-551-39-2582 > > > > > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.1 (GNU/Linux) > > iD4DBQFDaz8vUxlJ7aRr7hoRAg62AJ95spVCUiz1J/94bOlXp4khknhrMQCXfU5W > fid/FOukCnNLyh73vQpQKA== > =BgeK > -----END PGP SIGNATURE-----