SAD and MR-SAD phasing with the Phaser GUI
- Input files and mandatory parameters
- Frequently Asked Questions
Since the two Phaser GUIs use a common base, much of the experimental phasing
interface is similar to the Phaser-MR GUI. This section
only covers features unique to SAD phasing, including instructions for MR-SAD.
Phaser is used internally in the AutoSol wizard, which automates a complete
experimental phasing pipeline including heavy atom search, phasing, density
modification, and preliminary model-building. (It is also capable of running
MR-SAD experiments.) For that reason, we recommend trying AutoSol first,
and falling back to standalone Phaser for more difficult problems where finer
control over parameters is required.
This document only covers configuration of the Phaser GUI; for a detailed
explanation of the various terms and search procedures referenced here as well
as interpretation of results, consult the
documentation for AutoSol or Phaser, or the
Input files and mandatory parameters
Phaser requires a reflections file (any format) containing anomalous data,
and a file containing heavy atom sites, in either PDB or SHELX format. Either
the wavelength or the energy used is also required. To
run an MR-SAD experiment, change the mode to "MR-SAD phasing", and enter a
partial model (PDB file) and the expected variance from the actual structure
(RMSD or percent identity). If you are using normal SAD mode, you may choose
whether to use the given enantiomer ("hand") of the heavy-atom sites or the
inverse or both; in MR-SAD, the chirality of the partial model restricts the
sites to the given enantiomer.
The input for composition is identical to the Phaser-MR GUI.
Heavy atoms may be either standard elements from the periodic table, or
cluster compounds such as tantalum bromide or tungsten clusters.
Depending on the choice of hand, Phaser will output one or two sets of files
containing the heavy atom sites (PDB format) and phased reflections (MTZ),
including map coefficients and Hendrickson-Lattman phase distributions
(useful for refining with experimental phase restraints). This map is
usually very noisy and difficult to interpret, and typically requires
density modification to become useful.
Frequently Asked Questions
- How do I tell Phaser to look for more anomalous scatterers?
Click on the "Composition and sites" tab. A control at the bottom labeled
"Enable substructure completion by log-likelihood gradient maps" turns on
this feature (and is already on by default). However, Phaser also needs to
be told what type of scatterers to look for. Clicking the "Options..."
button opens a new window which will allow you to enter scatterer symbols
(typically element codes). You may enter as many of these as you think are
present and ordered in your crystal - note however that at the wavelengths
commonly used for macromolecular crystallography, some atom types may
be difficult to distinguish (such as sulfur and chlorine).
- How do I phase a structure with tantalum bromide clusters in Phenix?
This is possible, but not really optimal at present. Phaser has built-in
support for Ta6Br12 clusters, which have the scatterer type "TX". However,
they are represented as point scatterers rather than the actual cluster.
Therefore, although Phaser can perform LLG substructure completion, it will
output sites as single TX pseudo-atoms, rather than the actual cluster.