Dear all,
[just to add some caveats in case future students see this thread and
start cut-n-pasting commands to successfully create some kind of MTZ
file of dubious content]
Unless we're talking about some really old datasets (of historical
interest) or a disastrous situation (all backups lost), by far the
best approach would be to go back to the original data in the form of
the (unscaled) unmerged intensities - in order to do the
scaling+merging of the two datasets then, right?
One can do all kind of things with amplitudes (F) in MTZ files using
various fabulous CCP4 programs (SFTOOLS, SCALEIT, RSTATS, CAD,
POINTLESS, COMBAT), but be careful that it is not just some
simple/brainless arithmetic on numerical values that happen to be
called F/SIGF. Been there, done that, got the T-shirt ...
The important thing is to understand what the OP is trying to achive
here. As others mentioned:
... calculate R-merge for Fs from two datasets processed from two
different crystals ...
sounds to like a confusion between
(1) comparing two sets of amplitudes via an R-factor (not R-merge!)
* can be done with e.g. SCALEIT or RSTATS
* with or without scaling?
* what scale parametrisation (k, k/B, anisotropic)?
* overall or in bins? What binning?
(2) getting a "R-merge" value for some kind of "Table-1" (paper or
deposition):
* can't be done "correctly" from amplitudes
* go back to original data and (re)processing
And
... would take two mtz files and merge the Fs ...
might just ask for a method to "average" two sets of amplitudes:
* inverse-variance weighted?
* unweighted?
* for what purpose - and why not go back to the original data to do
proper scaling/merging there?
The subject line "Comparing two datasets" seems to ask for something
different again: a way of comparing (R-factor? CC?) datasets, not
necessarily combining/merging them?
Cheers
Clemens
On Tue, Jul 26, 2022 at 02:11:17PM +0100, Phil Evans wrote:
> If you want to merge them then you can use Pointless/Aimless/ctruncate, with
> the warning that the conversion from F to I is not ideal, and ctruncate
> should be set to regenerate F from I just as sqrt ie not apply the truncate
> procedure again.
> Pointless supersedes combat for this conversion
> Again, just for comparison use scaleit
>
> Phil
>
> Sent from my iPhone
>
> > On 26 Jul 2022, at 13:33, Jon Cooper
> > <0000488a26d62010-dmarc-requ...@jiscmail.ac.uk> wrote:
> >
> > Hello again, as far as I can tell, the question is about two already
> > merged/unique datasets which Mirek wishes to merge into one. As far as I
> > can tell/remember, Scaleit is for scaling datasets side-by-side to get
> > isomorphous differences, etc, and I don't know of a way that you could get
> > it to merge 2 datasets as Mirek requested. Probably I am wrong but Scaleit
> > R-factors are different from R-merge, although the latter might be a bit
> > dubious in the circumstances. Anyway, I think the Combat route is a viable
> > way of achieving what the questioner wants to achieve ;-0
> >
> >
> > Sent from ProtonMail mobile
> >
> >
> >
> > -------- Original Message --------
> > On 26 Jul 2022, 12:25, John R Helliwell < jrhelliw...@gmail.com> wrote:
> >
> > Dear Colleagues,
> > Scaleit is a terrific program.
> > Amongst its strengths already listed I would also mention its breakdown
> > table with F so that in the strongest F reflections bin any systematic
> > errors between the two data sets can show up if the Rfactor is greater than
> > about 1%.
> > Greetings,
> > John
> > Emeritus Professor John R Helliwell DSc
> >
> >
> >
> >
> >> On 26 Jul 2022, at 11:40, Eleanor Dodson
> >> <0000176a9d5ebad7-dmarc-requ...@jiscmail.ac.uk> wrote:
> >>
> >>
> >> Not only does SCALEIT do the job - it presented useful plots and an
> >> informative log file..
> >> Eleanor
> >> (You need to run CAD hklin1 Xtal1.mtz hklin2 Xtal2.mtz ...and obviously
> >> the columns from each Xtal will need different labels..)
> >>
> >> On Tue, 26 Jul 2022 at 09:45, Phil Evans <p...@mrc-lmb.cam.ac.uk> wrote:
> >>> If you give Pointless Fs, it squares them to Is (not correct if the Fs
> >>> have been derived from the truncate procedure, but not too bad). It will
> >>> then give you a comparison. If you give the two datasets to Pointless,
> >>> labelling them as different datasets, you can use Aimless to compare them
> >>>
> >>> But as Andrew said, the appropriate CCP4 program is Scaleit: it’s old but
> >>> still works and gives lots of statistics
> >>>
> >>> Phil
> >>>
> >>> Sent from my iPhone
> >>>
> >>>> On 26 Jul 2022, at 09:37, Andrew Leslie - MRC LMB
> >>>> <and...@mrc-lmb.cam.ac.uk> wrote:
> >>>>
> >>>> I think that POINTLESS works with intensities rather than structure
> >>>> factors (I’m not sure if this can be changed). Also, SCALEIT gives a
> >>>> much more detailed breakdown (R factors as a function of resolution and
> >>>> differences in terms of sigmas etc) than POINTLESS WILL.
> >>>>
> >>>> Cheers,
> >>>>
> >>>> Andrew
> >>>>
> >>>>> On 26 Jul 2022, at 09:24, LEGRAND Pierre
> >>>>> <pierre.legr...@synchrotron-soleil.fr> wrote:
> >>>>>
> >>>>> Hello Mirek,
> >>>>>
> >>>>> A very quick approach for that is offered by pointless:
> >>>>>
> >>>>> pointless HKLREF 1_1_aimless.mtz HKLIN 2_1_aimless.mtz
> >>>>> or
> >>>>> pointless HKLREF 1_1_aimless.mtz XDSIN XDS_ASCII.HK
> >>>>>
> >>>>> You will obtain a table looking like that, taking into account to
> >>>>> possible reindexing:
> >>>>>
> >>>>> Alternative indexing scores relative to reference
> >>>>> Alternative reindexing Lklhd CC R(E^2) Number
> >>>>> Cell_deviation
> >>>>> 1 [h,k,l] 0.993 0.962 0.118 19150
> >>>>> 0.08
> >>>>> 2 [-k,h,l] 0.007 0.078 0.512 19150
> >>>>> 0.87
> >>>>>
> >>>>>
> >>>>> Best wishes,
> >>>>>
> >>>>> Pierre Legrand
> >>>>> PROXIMA-1 Beamline
> >>>>> Synchrotron SOLEIL
> >>>>>
> >>>>> De: "Nicolas Foos" <nicf...@embl.fr>
> >>>>> À: CCP4BB@JISCMAIL.AC.UK
> >>>>> Envoyé: Mardi 26 Juillet 2022 08:36:35
> >>>>> Objet: Re: [ccp4bb] Comparing two datasets
> >>>>>
> >>>>> Hi Mirek,
> >>>>>
> >>>>> I am pretty sure XSCALE will do that for you :
> >>>>> https://xds.mr.mpg.de/html_doc/xscale_program.html
> >>>>>
> >>>>> If not, maybe have a look on SHELXC in SIR mode.
> >>>>>
> >>>>> Hope this help.
> >>>>>
> >>>>> Nicolas
> >>>>>
> >>>>> On 25/07/2022 21:52, Cygler, Miroslaw wrote:
> >>>>> Hi,
> >>>>> I would like to calculate the R-merge for Fs from two datasets
> >>>>> processed from two different crystals. Tried to use Blend but got the
> >>>>> message that Blend requires R. Downloaded R but do not know how to tell
> >>>>> CCP4 where it is located on my Mac. Is there another program that would
> >>>>> take two mtg files and merge the Fs?
> >>>>> Any help would be greatly appreciated.
> >>>>>
> >>>>>
> >>>>> Mirek
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>> To unsubscribe from the CCP4BB list, click the following link:
> >>>>> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
> >>>>>
> >>>>> --
> >>>>> Nicolas Foos PhD - ARISE fellow
> >>>>> https://orcid.org/0000-0003-2331-8399
> >>>>>
> >>>>> EMBL Grenoble, McCarthy Team
> >>>>> 71 av. des Martyrs,
> >>>>> 38000 Grenoble FRANCE
> >>>>>
> >>>>> +33 4 57 42 84 67
> >>>>>
> >>>>> To unsubscribe from the CCP4BB list, click the following link:
> >>>>> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
> >>>>>
> >>>>>
> >>>>>
> >>>>> To unsubscribe from the CCP4BB list, click the following link:
> >>>>> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
> >>>>>
> >>>>
> >>>>
> >>>> To unsubscribe from the CCP4BB list, click the following link:
> >>>> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
> >>>>
> >>>
> >>> To unsubscribe from the CCP4BB list, click the following link:
> >>> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
> >>>
> >>
> >> To unsubscribe from the CCP4BB list, click the following link:
> >> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
> >>
> >
> > To unsubscribe from the CCP4BB list, click the following link:
> > https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
> >
> >
> > To unsubscribe from the CCP4BB list, click the following link:
> > https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
>
> ########################################################################
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
>
> This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing
> list hosted by www.jiscmail.ac.uk, terms & conditions are available at
> https://www.jiscmail.ac.uk/policyandsecurity/
--
*--------------------------------------------------------------
* Clemens Vonrhein, Ph.D. vonrhein AT GlobalPhasing DOT com
* Global Phasing Ltd., Sheraton House, Castle Park
* Cambridge CB3 0AX, UK www.globalphasing.com
*--------------------------------------------------------------
########################################################################
To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list
hosted by www.jiscmail.ac.uk, terms & conditions are available at
https://www.jiscmail.ac.uk/policyandsecurity/
