[buster-discuss] Jubilee release of the Global Phasing software suite, with improvements and numerous fixes in BUSTER and Grade2

BUSTER Developers buster-develop at globalphasing.com
Wed Jun 8 12:42:35 CEST 2022


Dear BUSTER users,

     The Queen's Platinum Jubilee notwithstanding, we are pleased to
announce a new release of our software that includes numerous new features,
improvements and fixes in Grade2, BUSTER and Pipedream.

     The highlights below are only a small subset of the Release Notes, that
should be consulted for full details at

https://www.globalphasing.com/buster/ReleaseNotes/ReleaseNotes-BUSTER_snapshot_20220608.txt

     Note that due to the accidental omission of "aB_autorefine" in a limited
release made on June 3rd the relevant Release Notes appear under that date.

     All our programs have been checked for full compatibility with the
latest CCP4 8.0.001 release.


#### Highlights of improvements to Grade2.

     Grade2 will now recognise alpha amino-acid compounds and will ensure
that the main-chain atom naming follows standard PDB conventions (N, CA, C,
O, OXT and CB). This should simplify the use of such restraint dictionaries
for modified amino-acid residues as part of the macromolecule. The handling
of MOL2 input files has been greatly improved when it comes to partial
charges and bond typing. The interaction with Coot has been improved -
including a fix to the MMDB library itself (so newer versions of Coot should
no longer show problems with long InChi keys). To simplify fitting Grade2
into your particular workflow, the name of the output restraint file can now
be specified via the --ocif argument.


#### Highlights of improvements to BUSTER.

     A first version of a new automatic refinement tool ("aB_autorefine")
has been added to the distribution, that should make it even easier for new
users to get started with BUSTER refinement. This tool runs a series of
standard BUSTER refinements - from initial rigid-body to normal refinement
with TLS, occupancy and ADP refinement including hydrogens (if appropriate)
- and provides a graphical summary of all steps. By default, it will make
use of an experimental version of a variant of LSSR for NCS and includes
automatic rejection of log-likelihood outlier reflections.

     The method for automatic adjustment of (relative) X-ray weight in
BUSTER has been modified to use a resolution-dependent formula which should
result in a better balance between X-ray and geometry part. Three different
variants can be selected via -rpolicy flag : "loose", "medium" (default) and
"tight". In general, higher-resolution structures will by default refine to
a larger rms(bond) value (i.e. a higher weight to the X-ray term) while
lower resolution structures will maintain a tighter geometry.

     Further improvements in BUSTER include: 
     
     (1) better high-resolution refinement behaviour through the adjustment
of the bulk solvent mask blurring;

     (2) the refinement of large structures has been speeded up through
substantial code reorganisation and optimisation;

     (3) improvements in producing the PDBx/mmCIF file (BUSTER_model.cif)
for the description of the macromolecule as well as potential restraint
dictionaries;

     (4) a new macro ("AlternatingRefinement") for running alternating
parameter selection (XYZ, B and/or OCC) within a single BUSTER ("refine")
run;

     (5) better handling of C-terminal amidation;

     (6) further control options regarding occupancy refinement;

     (7) additional plots ("Plots/summary.png" and "Plots/summary_LL.png")
to summarise refinement behaviour.


#### Improvements to Pipedream.

     Rhofit solutions will now be sorted according to the values of the
correlation coefficient between the calculated electron density for the
ligand and the mFo-DFc difference map, since that has been found to be the
better performing selection criterion for the best solution.



As always, we kindly ask that support requests and feedback be sent to our
developer mailing list:

       buster-develop at globalphasing.com

rather than to individual developers.


     With best wishes,

The Global Phasing developer


More information about the buster-discuss mailing list