Dear Kyriacos,
Just one thing to add to this suggestion from Herman:
I personally view and study the plots generated by the SCALA/TRUNCATE or
AIMLESS/TRUNCATE route after XDS processing (in particular the plot of
Rmerge values as a function of Batch number) to eliminate deviating
batches from processing. Once I have done this, XDS/XSCALE/XDSCONV gives
me very adequate results (you will notice that I am not particularly
faithful to a single suite of crystallographic software but that I pick
what I want from each program suite... :-) )
HTH,
Fred.
On 21/03/13 08:41, [log in to unmask] wrote:
> Dear Kyriacos,
>
> What kind of high-resolution data do you have?
> In my experience, while Scala usually produces excellent results, it often fails miserably or even crashes due to too negative intensities in case of low redundancy data. E.g. if one does a second high-resolution scan with a high swing-out angle with little low-resolution data and friedel mates. I have not looked whether Scala could be stabilized by somehow restraining the scale factors or switching off some refinement parameters. Since we process with XDS, in these cases I use XSCALE for scaling, which does produce very good results.
>
> My recommendation would be to process with XDS. By using the autoProc procedure from Global Phasing this is very easy, even for people who are normally not able to run a program without a GUI. You will then have to run XSCALE manually, which is again trivial once XDS had run correctly.
>
> Good luck!
> Herman
>
> -----Original Message-----
> From: CCP4 bulletin board [mailto:[log in to unmask]] On Behalf Of Kyriacos Petratos
> Sent: Wednesday, March 20, 2013 6:25 PM
> To: [log in to unmask]
> Subject: [ccp4bb] Scaling with SCALA high and low resolution data sets
>
> Dear All,
>
> we have two data sets at about 0.9 and 1.9 Ang. resolution collected from a single crystal.
> Integration with iMosflm seems to be fine like the scaling within each of the data sets.
> When we try to merge and scale both of them with 'Scala' we get extremely high scale factors for the lower resolution images varying between approximately 30 and 200!
> Do we need to pay attention to some particular options for running the program(s)?
> Thank you,
>
> Kyriacos
> e-mail: [log in to unmask]
>
>
--
Fred. Vellieux (B.Sc., Ph.D., hdr)
ouvrier de la recherche
IBS / ELMA
41 rue Jules Horowitz
F-38027 Grenoble Cedex 01
Tel: +33 438789605
Fax: +33 438785494
|