MERGING THREE DATASETS

From: arka chakraborty
Date: 21 December 2011 11:15


Hi all,

I have three datasets  with same space-group and identical cell dimensions which I want to merge together. I remember that this was discussed in the blog some time back. Nevertheless, I can use some help!

Thanks in advance,

ARKO

-- 


----------
From: Eleanor Dodson


Do you still have the unmerged files?
In that case just feed them to pointless and redo scala/etc..

If you only 3 merged data sets, you will have to use combat to convert the mtz files into an fake "unmerged" format then put them through pointless/scala to get relative scales and merge them..

Eleanor

----------
From: Vellieux Frederic

There are several ways to skin a cat. If you have processed your data with XDS, XSCALE can also do the job (you have to think of the unit cell parameters of the merged data set though - the default may not necessarily what you wish to have), and I am certain other data frame processing suites will have this offered as an option somewhere... (I personally like all the statistics provided by the combat / scala route though).

Fred.

----------
From: harry powell


Hi

I'd have to say that I *really* like using Phil Evan's new program "Aimless" rather than Scala - not sure if it's in CCP4 yet but it is available from his own ftp site for a number of platforms -

       ftp://ftp.mrc-lmb.cam.ac.uk/pub/pre

(I think).

In my hands it's faster and more reliable most of the time.

So my route from unmerged files would be pointless then aimless...

Harry
--
Dr Harry Powell,

----------
From: arka chakraborty


Hi,

The solution provided by Mr. Dodson worked out well! Also it was nice to learn about the other options available.

Thanks a lot!

ARKO

----------
From: arka chakraborty

Hi,

would like to apologize for addressing Dr. Dodson incorrrectly...it was purely unintentional...

regards.
 
ARKO

----------
From: Yuri Pompeu


I was actually trying to do the same thing. I have to data sets processed in p21 that I would like to merge as one contains higher resolution data. They were processed and merged using the HKL suite. So I have the two .sca files
I guess the first step for me would be to create " a fake unmerged set" of both using Combat, as suggested by prof. Dodson and others.  Practically speaking, do I accomplish that by changing the space group to P1 while keeping my cell dimensions constant in combat?
Thanks a lot

----------
From: Eleanor Dodson 

If you still have the input to scalepack you can rerun it to get an unmerged file for both sets of data..

You dont need to change the SG unless of course there is a different indexing convention for the sets..

pointless should handle that - you can generate a mtz file for each data set, then ask to match the indexing to the first


Then you can combine the 3 mtz files, and get a single one ready for input to scala (or aimless)

..

Eleanor