------------ CCP4 Newsletter - January 1997 ------------

CCP4 - Recent changes

Martyn Winn & Adam Ralph

Daresbury Laboratory,
Daresbury,
Warrington
WA4 4AD, U.K.

Since the last Newsletter, version 3.2 of the CCP4 suite has been released. In addition to a number of bug fixes and enhancements to existing programs, this release includes the following new programs:

1. MAKEDICT: will produce TNT or PROTIN dictionary entries from a PDB or PROTIN file (Phil Evans).
2. MATTHEWS_COEF: written by Misha Isupov to calculate the Matthews coefficient given the cell, symmetry, number of atoms etc..
3. RASMOL: RasMol 2.6 (unix and VMS versions) is now distributed with CCP4 (Roger Sayle).
4. SCALEPACK2MTZ: Convert SCALEPACK output into an MTZ file. It then must go through TRUNCATE to convert to Fs.

A full list of changes for version 3.2 is appended below.

Documentation for most of the main programs is now distributed in html format, in addition to the man page source and formatted versions. Point your browser at $CCP4/html/INDEX.html These html files are at the moment generated automatically, and are perhaps not ideal. They will be gradually improved, and at some point the man page source files (man1/*.1 etc.) will be made obsolete.

An official mirror site for the Suite was set up in September at the Photon Factory, Japan by Atsushi Nakagawa. This supplements the other mirror site in San Diego. Details can be found on the CCP4 home page.

If you are having problems with CCP4 programs either compiling or running then have a look at the Problem Page. This contains various fixes effected since the latest release.

Finally, thanks to all those who visited us at the IUCr Commercial Exhibit Show in Seattle. We hope you were suitable impressed by the poster!

Changes to the Suite

Building etc.

• $CCP4/x-windows/XCCPJIFFY/Imakefile: compiler options for HPUX sent by Morten Kjeldgaard.
• $CCP4/x-windows/xdlgjk/Makefile: added install and empty-targets procedures, in line with $CPROG. Added MAKEFILE.COM
• $CPROG/MAKEFILE.COM: procedure for Solomon was causing a warning but compilation was not affected. This has been fixed to eliminate warning. Solomon compiled with case sensitivity.
• Compilation of C code changed due to library.h.
• configure: because of problems with the optimisation the default is now O1 for OSF1. IRIX6.2 section added. Warning messages have been surpressed for IRIX5*, HPUX A.09.* and OSF1.

Program changes

• AMORE: if the B for an atom is zero or below then it is reset to 20.0. A message is printed if this is the case.
• BONES2PDB: latest version from KDC on 7/10/96.
• DM: lastest version from KDC on 7/10/96.
• ECALC: resolution keyword is now acted upon.
• F2MTZ: calls to fatal were incorrect.
• FFT: keyword FILL didn't do anything - fixed. New keyword GRID SAMPLE to specify the sampling grid as a fraction of the resolution. FFTs in P2 did not work properly and has been removed. P1 should be used instead of P2.
• FHSCAL: labels DANO and SIGDANO changed to DPH and SIGDPH to be consistant with SCALEIT.
• IPDISP: spdfils have been changed for MARs. -Ms is now a bigmar with squashed format. -Mo has been removed but is equivalent to -M. -m small mar in original format. All these are big endian. -V is the same and now -v is small mar from VAX i.e. little endian.
• MAMA2CCP4: latest version from KDC on 7/10/96.
• MAPMASK: latest version from KDC on 7/10/96.
• MAPROT: latest version from KDC on 7/10/96.
• MLPHARE: problems with printing MNFs when monitoring reflections fixed.
• MTZ2VARIOUS: typo in defining external function was causing compilation problems. Now fixed. EXCLUDE FREER will now exclude any freeR subset. Any FreeR sub-set in the MTZ file can be output as the FreeR set to XPLOR, SHELX. Dummy columns IDUM?? have been added, these will output as integers in USER mode. An mmCIF file of the reflection data and associated data can now be generated. XPLOR output changed to output SIGMA instead of SIGM.
• MTZDUMP: FORMAT keyword introduced so that the reflection list can be output in a certain format. Resolution limits are applied to the overall statistics. Missing reflections are flagged with '?'. Ranges are now calculated for the partial statistics tables (STATS keyword).
• MTZUTILS: the UNIQUE option data from file 2 was output instead of file 1. Bug now fixed. Associated columns properly scaled with SCALE option.
• NCSMASK: latest version from KDC on 7/10/96.
• POLYPOSE: Bugs fixed so that default values work if FIX and INCLUDE not specified.
• POSTREF: the column label 'ABSFAC' and 'SIGABS' have been changed to 'SCALE' and 'SIGSCALE' in order to work with output from SCALA.
• PROCHECK: a write statement in PPLOT was over 132 characters. Changed the write to avoid this. Only a problem with VMS.
• PROTIN: +PROLSQ & REFMAC the format of the PROTOUT file has changed so that the files will not be compatible with last version. Change is so that residue numbers up to 9999 are allowed.
• REFMAC: Output FOM as well. Made P1 default symmetry. Now supports cubic spacegroups. Defaults have changed since last version please read the documentation.
• RESTRAIN: updated to version 4.3.5 from Ian Tickle. Isotropic Bs are written out properly when outputing aniso. Us.
• ROTAPREP: the way the batches are handled is now more consistant. NBATCH keyword has been removed, the number of batches is taken from the file where appropriate. MISBATCH keyword added to specify missing batches.
• SCALEIT: interesting stats. on large isomorphous/anomalous differences produced for REFINE and ANALYSE modes. Normal probability analysis now seperates centrics and acentrics. Now outputs Kraut scale instead of reciprocal.
• SIGMAA: problem with missing FC sorted.
• SOLOMON: CCP4 masks were not being read properly, now fixed. O style rotation and translation operators do not have to have the header line. Problems have have been caused when averaging using CCP4 masks, now resolved.
• TRUNCATE: divide by zero problem fixed. The output now contains H K L F SD Dano SD F+ SdF+ F- SdF- Imean SD I+ SdI+ I- SdI-. Also, scaling with the Wilson plot is compulsory.
• XDLMAPMAN and XDLDATAMAN: removed system dependent routines. Altered the menu and io windows so there are sized flexibly. Also, added command line qualifier so font for menu can be changed.

New programs

• MAKEDICT: will produce TNT or PROTIN dictionary entries from a PDB or PROTIN file.
• MATTHEWS_COEF: written by Misha Isupov to calculate the Matthews coefficient given the cell, symmetry, number of atoms etc..
• RASMOL: RasMol 2.6 (unix and VMS versions) are now distributed with CCP4.
• SCALEPACK2MTZ: Convert SCALEPACK output into an MTZ file. It then must got through TRUNCATE to convert to Fs.

Library

• $CLIBD/(no)chiral_pep*.dic: correction to ARG sidechain.
• $CLIBD/atomsf_neutron.lib: Change atomic number of D to 101 to bring into line with RBROOK.
• chelp.c: some irregularities fixed.
• chelp.h: machine dependent stuff now in library.h.
• library.c: integer type was incorrectly defined for Big-endian machines with CONVERT_FROM set.
• library.h: has been created from library.c. This will contain the machine dependences required for all C programs.
• mtzlib.f: bug in LRREFL fixed; affected BIOMOL programs.
• rwbrook.f: s/r SFREAD moved here from AMORE,MLPHARE,REFMAC,SFALL

Documentation

• RSTATS: document brought in line with prog. FREE keyword added.
• Tutorials: a new set of tutorials have been made in five main areas: MIR, MR, refinement, density moification and MAD. These procedural scripts can be found in $CEXAM/tutorial.
• html versions of the documentation for most of the main programs are in $CCP4/html Point your browser at $CCP4/html/INDEX.html
• x-windows/doc: file names have been changed but essentially the *.txt files are similar to the old *.doc* files.

Withdrawal

• ROTAVATA: will be unsupported in release 3.3. SCALA will become the recommended program.
• PROLSQ: will be unsupported in release 3.3. REFMAC will become the recommended program.

Back to Contents....