>On Feb 28, 2007, at 14:37, shivesh kumar wrote:
>
>Dear all,
>I have a data set at 2.2A, of the selenomethionene labelled
>protein.How should I process the data.
Some hopefully useful remarks (fairly random and not complete and
exhaustive):
1. make sure to mask out the backstop and beamstop holder
correctly. Although various integration software claims to do this
fairly automatic it is always better to do a good job on this.
2. check the rejected reflections (at the scaling/merging step): is
there some system in those rejections? The two files produced by
SCALA (ROGUES and a xmgr file that plots the detector position of
rejected reflections) are very helpful. It can show ice-rings, bad
beamstop-masking (see point 1) etc.
3. heavy atom detection/phasing software will also write out some
helpful information about outliers: if there are some suspicious
messages (e.g. warnings in autoSHARP) they usually point back to
problems in data processing (see point 1).
4. if you collected several datasets/wavelengths: you can give those
to SCALA for some 'local scaling'. This will also show you those
really helpful CC(Dano) plots. But be careful with absorption
correction: all datasets/sweeps need to be indexed identically.
5. always remember what your first reaction was when looking at the
images: 'great' images should give you good statistics further down
the pipeline. But if 'awfull' images give you good statistics I
would be suspicious ...
6. unless you really know what's going on: stick with program
defaults. Usually the developers have a very good idea why the
program is doing things in a specific way.
Cheers
Clemens
--
***************************************************************
* Clemens Vonrhein, Ph.D. vonrhein AT GlobalPhasing DOT com
*
* Global Phasing Ltd.
* Sheraton House, Castle Park
* Cambridge CB3 0AX, UK
*--------------------------------------------------------------
* BUSTER Development Group (http://www.globalphasing.com)
***************************************************************
|