Thanks for the replies Harry, Joe and Kay,
@Harry
(1) I've had a look at the com files and the the run with the file from XSCALE reaches the ROGUEPLOT stage (command line) and says mtz MERGED (Script from file), when I press continue from that stage it switches from running to failed. The run with the ASCII file before running XSCALE continues on normally to completion.
(2) Yes both the command line and ccp4i are running Aimless 0.1.29.
@Joe The XDS_ASCII.HKL that is running through Aimless on ccp4i and finishing is from CORRECT. I've tried running the output from XSCALE through pointless using the commands you suggested and Aimless is still giving out the same error message on ccp4i and on the command line.
@Kay I had a look at the output from XSCALE and it has "!GENERATED BY: XSCALE (VERSION July 4, 2012)". I'm using the computer systems at Diamond so don't think I'm able to update this easily, I still don't really know how the permissions work here. From my relatively inexperienced perspective, the statistics look very similar between XSCALE and AIMLESS but the main reason for running through Aimless after XSCALE is it gives a value for Rp.i.m. which I don't see in the XSCALE.LP file. Unless it is under a different name somewhere?
Thanks again,
Sam
________________________________________
From: CCP4 bulletin board [[log in to unmask]] on behalf of Kay Diederichs [[log in to unmask]]
Sent: 26 July 2013 17:49
To: [log in to unmask]
Subject: Re: [ccp4bb] Processing files from XSCALE in Aimless
On Fri, 26 Jul 2013 15:19:57 +0100, Joe.Lyons <[log in to unmask]> wrote:
>
>Hi Sam,
>
>Assuming the XDS_ASCII.HKL is from XSCALE. I generally run it through Pointless then aimless with (a scales constant command) using something like the following.
>pointless -copy xdsin XDS_ASCII.HKL hklout sorted.mtz << eof > pointless.log
>eof
>
>aimless hklin sorted.mtz hklout Scaled_1.mtz << eof > aimless.log
>run 1 all
>scales constant
>resolution 60.00 4.00
>anomalous on
>eof
>
>If the reflection file is from the CORRECT step I would remove the scales constant command when running aimless.
I do hope you know that CORRECT scales the data, and when you "remove the scales constant command " then you scale them again in aimless. Are you certain that this is good for the signal in your data?
best,
Kay
>
>Hope this helps,
>
>Regards,
>
>Joe
>
>-----Original Message-----
>From: CCP4 bulletin board on behalf of Horrell, Sam
>Sent: Fri 7/26/2013 1:32 PM
>To: [log in to unmask]
>Subject: [ccp4bb] Processing files from XSCALE in Aimless
>
>Hi Everyone,
>
>I've been trying to process some data using Xia2, XSCALE, Pointless and Aimless. I have a large number of datasets (over 100) so have used Xia2 to produce the XDS_ASCII.HKL files using the 3daii pipeline. I now want to process them manually (with a script) using XSCALE and pointless to find the resolution limit and finally run them through Aimless to get the merging statistics. I'm not looking to solve the structures just to compare the merging statistics between datasets subjected to different treatments.
>
>I've managed to put the the XDS_ASCII.HKL files through XSCALE and pointless without any problems but when I try and run aimless with the output from XSCALE in CCP4i I get the error message
>
>"FATAL ERROR message:
>
>ERROR in ScaleModel: run 1 has insuffient information for smooth scaling"
>
>And when trying to run it by command line Aimless won't recognize the XSCALE.HKL file.
>I've tried converting the XDS file with XDSCONV to the ccp4 format but Aimless still didn't like it on command line or ccp4i so I tried using sortmtz but this just fails with a very uninformative message
>
>
>"SORTMTZ: *** ERROR: Program Terminating
>
>Times: User: 0.1s System: 0.0s Elapsed: 0:01"
>
>The command line I've been trying to use for aimless is
>
>"aimless hklin XSCALE.HKL hklout scaled.mtz "
>
>Running Aimless with the XDS_ASCII.HKL from Xia2 using ccp4i works fine.
>
>Is there a conversion step I'm missing when trying to run the XSCALE.HKL files through Aimless?
>Maybe something I've done wrong in the XDSCONV stage or I need to get sortmtz working somehow?
>
>Any advice would be greatly appreciated.
>
>Best regards,
>
>Sam
>
>
>
>
>
|