Dear All,
An additional comment:
Analysis is built so that it (re)calculates shifts from the positions of
peaks they are assigned to. I once used to control the precise frequencies
to use carefully one by one, and was sceptical of this averaging business
at the start, but it does save you a lot of work. Setting the weight of
all spectra to zero should nullify the averaging.
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
On Tue, 6 Oct 2009, Wayne Boucher wrote:
> Hello,
>
> Tim says (interspersed below):
>
>> Date: Tue, 6 Oct 2009 13:49:47 +0100
>> From: Lalit Deshmukh <[log in to unmask]>
>> Reply-To: CcpNmr software mailing list <[log in to unmask]>
>> To: [log in to unmask]
>> Subject: few questions
>>
>> Hi,
>> 1)I have almost 90% complete chemical shift table (in CYANA format,
>> generated by CARA). I import it to analysis using format converter which
>> does a decent job except for Val and Leu residues. The methyl groups
>> (CD1,CD2,CG1,CG2) are imported non- stereospecifically , thus when I create
>> synthetic peak list, analysis will create the peaks for all combinations
>> (eg: CD1 will be paired with both HD1 and HD2). Is there a way around this
>> ?
>
> This may be more of a Wim question, but prochirals can be resolved by
> assigning them stereospecifically if needs be. For non-stereo assigments the
> covalent links, which would pair the correct H-C, come from assignments in
> spectra (and knowledge of the 'onebond' dims). If there are no spectrum
> assignments this cannot be done.
>
>> 2) All my spectra are calibrated for my chemical shift table, however when
>> analysis will create a synthetic peak list, it moves the shifts around.
>> Although there are no huge differences , I would like to have a fixed
>> chemical shift table ( as in case of PIPP or CARA, both the softwares will
>> not change the shifts unless and until I want to change them). Is there
>> some
>> way where the shifts will be fixed and synthetic peak list will be
>> generated
>> based upon that fixed set of resonances?
>
> The synthetic peaks are severely down-weighted so that they have almost no
> influence on shifts compated to normal peaks. The factor is 0.0001 from
> memory. They have some influence so that it is still possible to derive
> shifts from them (e.g. put in a shift list on their own a value is recorded).
>
> I guess the problem comes when you have no real peaks in CCPN, which isn't
> the case for most people, which would hold the shifts firm. A potential
> solution is to set Edit Spectrum:Tolerances - Shift Weighting to zero for the
> spectra that the synthetic shifts are made for.
>
>
>> 3) While generating a synthetic peak list for NOESY spectra, can I specify
>> in advance the threshold for the particular spectrum ( for example in
>> Auremol, the software will calculate the threshold either automatically or
>> manually, and pick the peaks above this threshold)?
>
> Easy enough to add to the to-do list, but hopefully deleting peaks afterward
> based on intensity should suffice functionally. [Wayne adds: you can sort on
> intensity in the peak table by clicking on the header.]
>
>
>> 4) While generating the synthetic peak list (say based on the structure)
>> lot
>> of peaks are picked multiple times, is there a simple way to remove the
>> duplicate peaks or merge them?
>
> I suspect it will be best to stop the duplication in the first place. I do
> not have time at the moment though.
>
|