Dear Theresa
In response to your first question, one way to do this is to look
closely at the image (eg changing the contrast level) with the
predicted pattern superimposed after indexing, or alternatively just
integrate the image and look at the reported mean I/sigmaI as a
function of resolution. Both of these are trivially easy to do with
iMosflm.
You need to remember that any number of issues (eg zingers, hot
pixels, ice diffraction) can give rise to what are apparently
significant intensity spots, so to distinguish between "real
diffraction" and artifacts you need to look carefully at the spot size
and shape as well as its intensity (and whether it is predicted).
The other point to bear in mind is that generally there is no problem
integrating the images to a resolution a bit beyond where spots are
visible, eg to 3.0-3.2Å when spots are only visible to ~3.4-3.5Å and
then using the merging or refinement statistics to decide the "true"
resolution of the data.
Best wishes,
Andrew Leslie
On 8 Oct 2012, at 23:36, Theresa Hsu wrote:
Dear all
I took some images from test crystal and tried to index them to get
cell parameters, not enough for structure solution. I can see spots
at 3.5 A with ADXV but when I indexed, XDS reports up to 3.0 A in
INCLUDE_RESOLUTION_RANGE in XDS_ASCII.HKL.
Questions:
1. How do I know that the spot at 3.0 is not background noise?
2. What is the function of FRAME.cbf file?
Thank you.
Theresa
