Dear Bulletin Board,
Thank you very much for the information and comments. In the last days I have learned a lot and it has been very helpful. It turned out that at the basis my problems were the fact that sftools was not up to date and that I had been looking in the wrong places for space group names:
Intl tables vol A for space group information -> should be intl tables vol G (see below)
Symop.lib for ccp4 symmetry information -> should be syminf.lib
Syminf.lib indeed seems to contain all information needed: four different names for space groups: "ccp4", "Hall", "xHM" and "old", where "old" seems to correspond to the PDB definition and "xHM" to the definition used by phaser. Since I was trying to fix the space group name for a process (autobuster) which was also using sftools, one starts going around in circles. Knowing this, I went back to cad and this program did indeed correctly replace the space group name "R 3 :H" by "H 3", so I assume that all ccp4 programs which use the official subroutines will correctly recognize alternative names for space groups.
Best regards,
Herman
-----Original Message-----
From: CCP4 bulletin board [mailto:[log in to unmask]] On Behalf Of Miller, Mitchell D.
Sent: Thursday, September 02, 2010 6:31 PM
To: [log in to unmask]
Subject: Re: [ccp4bb] Sftools and Phaser compatibility issues - continued
FYI --
The :h suffix for R3 is described in the IUCr symmetry cif (intl tables vol G chapter 4.7) under _space_group.reference_setting where it states "For the space groups where more than one setting is given in International Tables, the following choices have been made. For monoclinic space groups: unique axis b and cell choice 1. For space groups with two origins: origin choice 2 (origin at inversion centre, indicated by adding :2 to the Hermann-Mauguin symbol in the enumeration list). For rhombohedral space groups: hexagonal axes (indicated by adding :h to the Hermann-Mauguin symbol in the enumeration list)."
http://it.iucr.org/Ga/ch4o7v0001/ch4o7.pdf
http://www.iucr.org/resources/cif/dictionaries/cif_sym
The H3 / H32 designations are PDB conventions/standards. In the PDB format description it states that "For a rhombohedral space group in the hexagonal setting, the lattice type symbol used is H."
From an archive of the PDB documentation at the University of Washington, there is list of changes by PDB version that suggests that the PDB introduced the H designation with the release of PDB format v2.0 (sometime around March 1997) see http://www.bmsc.washington.edu/CrystaLinks/man/pdb/guide2.2_frame.html
http://www.bmsc.washington.edu/CrystaLinks/man/pdb/part_6.html
The RCSB's archive of the 2.2 format gives a file not found error.
http://www.rcsb.org/pdb/docs/format/pdbguide2.2/guide2.2_frame.html
Regards,
Mitch
-----Original Message------
From: CCP4 bulletin board [mailto:[log in to unmask]] On Behalf Of Nat Echols
Sent: Thursday, September 02, 2010 8:06 AM
To: [log in to unmask]
Subject: Re: [ccp4bb] Sftools and Phaser compatibility issues - continued
On Thu, Sep 2, 2010 at 5:31 AM, <[log in to unmask]> wrote:
The other question is: why does phaser write 'R 3 :H' in the mtz? When the problem with the P21221 space group first popped up last year, Randy told me that space group numbers like 2018 are non-standard, and that space group 18 with the name P21221 was the way to go. This is fair enough, but 'R 3 :H' is neither PDB nor ccp4 standard and I did not find it in the international tables. Is it maybe a phenix standard?
No, it pre-dates Phenix - it's the "extended Hermann Mauguin symbol", whatever that means:
http://www.ccp4.ac.uk/html/symmetry.html
I don't know why it's used preferentially in Phenix, but in theory it's supported by CCP4 programs, except those which are still using the older symmetry information. syminfo.lib has the correct information (space group number 146), symop.lib does not. As previously noted the last time this discussion came up (December, if memory serves), Coot also uses this notation:
http://www.biop.ox.ac.uk/coot/doc/coot/Reading-coordinates.html
-Nat
|