Uppsala Software News @ 970715 ****************************** http://xray.bmc.uu.se/~gerard/manuals *************************************** * SPANCSI Version 0.2 ********* This is a new program (part of the RAVE package). SPANCSI can be useful if you have indications that (or want to investigate if) your NCS-related molecules have very different average temperature factors, for example due to differences in the packing. SPANCSI allows you to: - calculate density statistics for each of your NCS-related molecules individually - calculate correlation coefficients etc. for each pair of NCS-related molecules - average and expand density using variance weighting for each of the NCS-related molecules. For example, suppose you have two molecules for which the variance of the molecular density is 100 and 50, respectively. Now in the averaging, the contribution of the second molecule will be up-weighted by a factor of 100/50 = 2.0, and when the averaged density is expanded back into the unit cell or asymmetric unit, the density for the second molecule will be down-weighted by a factor of 50/100 = 0.5. This hopefully leads to better convergence and better phases and maps. If you find that there are differences between the molecules, change your averaging script so it uses SPANCSI instead of AVE. * HETZE Version 0.3 ******* This is a new program (part of the X-UTIL package) to check the quality of hetero compounds, either prior to or after refinement. See the manual: http://xray.bmc.uu.se/~gerard/manuals/hetze_man.html . If you can't be bothered to download the program, you can use the HETZE WWW server at: http://xray.bmc.uu.se/~gerard/servers/hetze.html . (Note: if you do download, don't forget to download the required library file hetze.lib from pub/gerard/xutil/xutil_etc/ !) * XPLO2D Version 2.6.1 ******** Loads of changes (mostly improvements ;-) to the AUTO option to generate X-PLOR dictionaries: - the automatic "guestimation" of the number of hydrogen atoms attached to C, N, O, and S atoms has been vastly improved. The program gets it right most of the time now without human assistance (not always, of course). - the program can now use multiple copies of a compound so as to get better statistics for the bond lengths, angles, dihedrals and impropers. Just put multiple (up to 100) copies of your hetero compound into one PDB file and feed it into the program ! Beware that people may have different atom naming conventions and may have used wildly different sets of target values in their refinement. - the program also uses a different criterion to determine if two atoms are bonded, namely if their distance is no greater than the sum of their covalent bond radii plus a small tolerance (0.45 A). - all atom types with 3 or more non-hydrogen neighbours now give rise to an improper to keep them flat, or tetrahedral, etc. - "unusual" flatness and chirality impropers are automatically reset to standard values (0, +35 or -35 degrees). - there is a new option to assign every atom its own type (for high resolution input models, or when type conflicts occur, e.g. manifested through large angle ranges). - every bond now gives rise to no more than one DIHEdral (to prevent clashes and redundancies). - what constitutes a "large" observed range for bond lengths etc. is now defined by the user. - the X-PLOR minimisation job now also lists any close contacts. Note that the order of the input items has changed slightly. This means that a new version of the "het_dicts" script had to be made (can be downloaded from the OMAC directory). Also, if required, default file names and numerical parameters can be used so as to reduce the amount of input. New dictionaries have been generated for more than 1,000 compounds (but using version 2.6 rather than 2.6.1 of the program). These are accessible via the "hetero" page on the home page of the Uppsala Software Factory, or directly at: http://xray.bmc.uu.se/xhet/ . The XDICT server also uses the latest version of the program, and the input has been reduced to the bare minimum: http://xray.bmc.uu.se/~gerard/servers/xdict.html . * START-UP MACROS ***************** A number of programs now support start-up macros, i.e. macros that are executed automatically when a program starts up. This is useful, for instance, if you always work with X-PLOR files of nucleic acids in LSQMAN (set the central atom to P and the chain-mode to XPlor). or to set your standard grid and cell in MAMA (using the NEw commands). The feature works by setting an environment variable to the name of the file which contains your start-up macro. It is available in: - MAPMAN version 5.1 (setenv GKMAPMAN macrofile) - MAMA 4.1 GKMAMA - DATAMAN 5.1 GKDATAMAN - LSQMAN 5.4 GKLSQMAN - MOLEMAN2 1.2 GKMOLEMAN2 - ODBMAN 1.6 GKODBMAN * MAMA Version 4.2 ****** The ATom_fit command now checks both ATOM and HETATM atoms. In addition, there are two new optional parameters, namely a PDB file for all atoms inside the mask, and a PDB file for all atoms outside the mask. You could use this to find all internal water molecules in your model, for example (suggested by Thomas Simon). Made the ouput of the OVerlap TRim command a bit clearer when it checks to see if you specified an asymmetric unit plus 1, 2 or 3 points on all sides (thanks to Jan Dohnalek). * MAPMAN Version 5.3 ******** New PEek commands to extract or integrate density at/near atomic positions (added on request). There are three commands: - PEek VAlue - extract value (nearest grid point, or linear or spline interpolation) for each atom - PEek CUbe - integrate (absolute) density (or calculate rms value) in a parallellopiped around each atom - PEek SPhere - ditto, but in a sphere around each atom The calculated value is stored in the B-factor column of each atom, and a new PDB file is created. The values can be used to colour your molecule in O, or averaged per residue and plotted with MOLEMAN(2), etc. The MAppage and BRix commands have been changed so that your DSN6/BRIX files can now have more than 32767 records ... (this was a relic from the FRODO/PDP11 era, in which only 32767 records could be addressed by an integer*2; this number was still hanging around in the OPEN statements as a "MAXREC=" qualifier ...). From this version on, BRIX maps can also be read back into MAPMAN (again, on request) (use: REad map filename brix). Added a map format type "XPLOR" which is easier to type than "X-PLOR" ;-) * DATAMAN Version 5.1.2 ********* Removed bug from the EVen_kill command (reflections with a negative index were not treated correctly). Better estimates of unique reflections for some lattice types (thanks to Zhongning Yang). * LSQMAN Version 5.4.2 ******** Removed small bugs which under exceptional circumstances led to wrong results for ALL atoms, NONHydrogens and SIDEchain atoms. Improved statistics summary for MRama, MDihe, MSide and MTors commands (thanks to Pedro Matias). * MOLEMAN2 Version 1.2.2 ********** Checks for "weird" B-factors and occupancies while reading a new PDB file. Implemented X-PLOR polars and X-PLOR/Lattmann Euler angles in XYz ROtate (see: J Appl Cryst 30, 402-410, 1997). * OMAC ****** See: http://xray.bmc.uu.se/~gerard/manuals/omac_index.html Recent changes and additions: - het_dicts/hetero.pdb - new verion - multi_het_dicts - same as het_dicts, but extracts all examples of a hetero compound from structures solved at 2.0 A resolution or better - average.csh - updated for current CCP4 (thanks to Tom Taylor) - cheap_dejavu.csh - script to align your model to any or all structures in the PDB (uses the BRute_force command in LSQMAN; takes days !) - different_radii.odb - version of .cpk_radii in which the radius increases with each row of the periodic table (by Aaron Oakley) - align_many.csh - script to align a load of molecules to one other (uses LSQMAN) - rs_rfree_zone.inp - X-PLOR input file to generate systematic SA omit maps for your whole molecule (e.g., 10 residues at a time); you may also want dsn6.csh, a script which generates (1) a MAPMAN input file to mappage all the systematic omit files, and (2) an O macro to go through the results (and to calculate "real-space Rfree" values, i.e. real-space fit values inside each residue's own omit density) --cd (aka: http://xray.bmc.uu.se/~gerard/cd.jpeg ) CD's Word of the Month: Omphaloskepsis