[buster-discuss] Temp factors for solvent refining to very low values?

[Clemens Vonrhein]

Hi Ashley,

On Fri, Apr 21, 2017 at 01:27:23PM +0000, Ashley Pike wrote:
> Thanks for taking time to respond (the first response to this
> enquiry!). For info - got to the bottom of the problem. Only
> appeared to occur when data processed in a certain way - odd
> behaviour observed with DIALS processed data (DIAMOND auto pipeline)
> and/or data run through STARANISO truncation process.

Was that DIAMOND/DIALS processed data then given to the STARANISO
webserver? Or something else?

Depending when you collected the data, our processing pipeline
autoPROC [1] is also available and run at DIAMOND. Or you could
re-process the data via autoPROC there - which will include the
STARANISO analysis of anisotropy as well.

Of course, if there is something peculiar about data processing (with
one or several programs/pipelines), there could be something
problematic with the crystal, the data collection or the processing
itself. The output from autoPROC (summary.html) can be very useful in
highlighting those issues [2].

> Dataset is mildly anisotropic (3.1 x 3.7A) so suspect something
> going wrong with scaling/compensation carried out by these
> processes.

Scaling wouldn't really completely compensate for that: if
the crystal diffracts anisotropically the correction applied e.g. in
STARANISO [3] could help yes. What do the plots from STARANISO look
like?

> Solution was to simply re-process (XDS then aimless)

That is what autoPROC would/will do by default as well.

> and resulting dataset gives OK results (no atoms with B's at 3
> although some are or the lower than expected side).

Another diagnostics plot comming out of BUSTER is the Reciprocal Space
Correlation Coefficients Plot[4]. Does that give any indications of
problems with the data and/or model?

You can find those plots in the buster-report output (or when running
"graph_autobuster_recipCC" on the command-line).

> PS. Waters are 'high confidence' - manually placed based +fofc peaks
> and environment although there is no higher resolution structure to
> fall back on to confirm existence for sure.

Sounds like a sensible approach, yes.

Cheers

Clemens, Claus & Gerard (for buster-develop)

[1] http://www.globalphasing.com/autoproc/
[2] http://www.globalphasing.com/autoproc/manual/autoPROC7.html
[3] http://staraniso.globalphasing.org/
[4] http://www.globalphasing.com/buster/wiki/index.cgi?BRrecipCCplot