Hi,
I have a 2D spectrum, all peaks assigned but only on one side of the
diagonal.
Now I want to assign the other side of the diagonal, swapping
dimension = > F1 should become F2 and vice versa.
I cannot to it in > Peaks : Peaklist : = no option to change F1 and
F2, is there?
So, I exported the peaklist (I chose xeasy format), swaped x and y in
the .peaks-file and then imported it again following the "strategy"
Rasmus proposed to Matthias below.
The peaks are imported correctly, but there is NO assignment displayed!
If I import all 3 (.peaks, .prot, .assign), nothing changes. what to I
do wrong?
Carolin
On Jul 15, 2011, at 5:10 PM, Rasmus Fogh wrote:
> Dear Matthias,
>
> First the solution, then some comments.
>
> To import assigned peaks from CYANA, with your test files:
>
> - Open FormatConverter
> - Go to M:Import:combined files:shifts and peaks:XEASY
> - Set CYANA format on
> - select peak file
> - Select shift file
> - Do *NOT* select a peak assignment file
> - In advanced options:
> - set 'Use existing Data Source' to the approriate spectrum
> - leave 'Match existing Resonances' on
> - Set 'Minimal Interactions
>
> - Press the button to import
> - You will be asked how to map peak list dimensions to experiment
> dimensions. Leave as is. If that gives the wrong result try again
> and swap the dimensions.
> - When the program asks you if you want to run 'linkResonaces' say
> *no*.
>
> You will now have an extra imported peak list, with assignments.
> There will be no extra shift lists.
>
> To use structures to check unassigned NOEs:
>
> - Go to M:Experiment{Experiments}
> - Double click the 'Mol Systems' column, and set the structure you
> want to On.
> - Now you will get a structure set when you go into the assignment
> dialogue
> - Note that you only get distances showing when one of the peak
> dimensions is assigned.
>
> Normally I think this is set automatically. I do not know why you
> need to do it manually in this case.
>
> ========================================
>
> Some comments:
>
> Reading the .assign file gives you an error, something about
> 'invalid literal for int() 'Corresponding'. Basically the .assign
> files you have are totally wrong for the XEASY .assign file reader
> in FormatConverter. Anyway CYANA keeps the assignments in the peaks
> file, so you do not need them.
>
> Apparently, linkResonances ot its equivalent is done automatically
> when reading CYANA peak files.
>
> As you see, you can get the right result once you know the sequence
> of steps to use. Finding the right steps can sometimes be difficult,
> as in this case. FormatConverter was written to cover a *lot* of
> formats with a generic approach and a minimum of special-case coding
> for each, and it does its job. We are working on a replacement for
> FormatConverter, FormatExchange, that will use the tested file
> parsers of FormatConverter, but add an interface that is smoother,
> does more for you, and is easier to get to grips with. It is a lot
> of work to write, though, so it will be quite some time before it
> covers all the formats tha FOrmatConverter does.
>
> Yours,
>
> Rasmus
>
> ---------------------------------------------------------------------------
> Dr. Rasmus H. Fogh Email: [log in to unmask]
> Dept. of Biochemistry, University of Cambridge,
> 80 Tennis Court Road, Cambridge CB2 1GA, UK. FAX (01223)766002
|