Dear Sarathy,
I was going to reply along the same lines as Kay Diederichs just
did. At the Cr K-alpha wavelength, you would have strong absorption
effects which, given the low symmetry of your crystal, would be very
hard to get rid of. We have seen numerous cases where a high anomalous
correlation was spurious, in the sense that it did not correspond to a
solvable substructure but was very likely to be caused by systematic
errors, such as absorption effects or correlations in radiation-damage
effects, not removed by the scaling procedure.
This can be the big drawback of having a single axis of rotation
and low symmetry, as you do here. In such cases, collecting images
with at least two crystal orientations by means of a multi-axis
goniometer (or from several isomorphous crystals, counting on getting
a reasonable sampling of possible orientations just through chance)
would enable you to check to what resolution you get consistent values
for anomalous differences between orientations or between crystals.
This would be a much more significant indicator that the CCanom value
within a single dataset. Characteristically, your CCanom values are
more or less constant as a function of resolution, which is not a
feature of a true anomalous signal.
Together with the argument put forward by Richard Gillilan on the
basis of the high extinction coefficient of methylene blue, I am not
optimistic about the prospect of an izit-based S-SAD solution of your
structure - but would of course be delighted if you proved this to be
incorrect by actually solving you structure from this dataset!
With best wishes,
Gerard.
--
On Fri, Nov 30, 2012 at 05:27:41PM -0500, Sarathy Karunan Partha wrote:
> Dear all,
>
> Here is the XSCALE.LP output after scaling for the izit dye stained
> crystak. Sorry for attaching the .INP file.
>
> Sarathy
>
> On Fri, Nov 30, 2012 at 4:19 PM, Sarathy Karunan Partha <
> [log in to unmask]> wrote:
>
> > Dear all,
> >
> >
> >
> > We did some Izit dye staining to test our crystal (salt or protein) and we
> > observed that the crystal didn’t take up the dye well. But, showed nice
> > protein diffraction (home source KCr 2.2909 A) and we collected a dataset
> > (360 frames, 1o osc, 5 min exposure) on this dye stained crystal. The
> > data collection statistics looks great and most interestingly we saw some
> > anomalous signal for this data (see attached XSCALE.LP).
> >
> >
> >
> > This protein was crystallized in 0.2 M Ca acetate, 30 % PEG 400 pH 4.5.
> > Izit dye is basically methylene blue which contains a sulfur atom
> > (phenothiazine ring) and also has some basic dimethylamio groups. Our
> > protein has many acidic residues that could enhance binding of this basic
> > dye.We think the anomalous signal could be from this dye and the heavy atom
> > search with autoSHARP and Phenix autosol is suggestive of 4 sulfur atoms.
> >
> >
> > Did anyone come across similar situation with using this dye and also
> > welcome any suggestions about using this data for S-SAD phasing.
> >
> >
> > Thanks,
> > Sarathy
> >
--
===============================================================
* *
* Gerard Bricogne [log in to unmask] *
* *
* Global Phasing Ltd. *
* Sheraton House, Castle Park Tel: +44-(0)1223-353033 *
* Cambridge CB3 0AX, UK Fax: +44-(0)1223-366889 *
* *
===============================================================
|