> Couple more things to clear up:
>
> - Error of the fit is presented as an error of the relaxation TIME (note
> RATE) ?
Ah, I think I finally see what the issue here is.
The error when grouping peaks and fitting graphs is the error associated
with the differences between the experimental intensity values and the the
fitted function, as it should be. It is nothing to do with how accurately
intensities are measured, but rather a measure of goodness of fit. The
idea is that you can discover for which groups the fitting went poorly.
However, I see that it is inappropriate for the same error to be carried
forward to the measurement list (T1 or whatever) where it appears as if it
is an error in the time constant. The error that is needed here is
different. Assuming a good function fit, what we need is the variation in
the time constant for various fits due to the error in both time and peak
intensity.
I will discuss this with Wayne and we may be able to put something in for
the next release.
> > Error values can be entered for the time points of an experiment series,
> > so maybe in the future we could use these in the error for the overall
> > function fit.
> - So the error values entered by the user into experiment series are
> have no effect for now.
Indeed.
> Still which variable do these errors belong to -
> TIME or INTENSITY ?
Time. Time is what the condition points of the series relate to.
Although there is a slot in the Data Model for the error in peak intensity
of every peak it is not currently filled in (or even accessible from the
Analysis gui).
> What would make our life really easy would be an ability to calculate this
> uncertainty from crosspeaks in two given spectra, collected with the same
> time delay.
> You could nominate two spectra to estimate error for a time series ?
I'll look into this. Shouldn't be too hard to do.
> A possibility of entering user provided uncertaintity in the
> measurement of the intensity in an experimental series. Which would be
> estimated outside of Analysis and override any other error estimation ?
I think the order of precedence would be:
Use the intensity error for an individual peak in the group, if set.
Use the intensity error for the reference peak of each peak group, if set.
Estimate the error for a group based upon two spectra recorded at the
same time delay.
Guess a global intensity error from the noise.
Tim
-------------------------------------------------------------------------------
Dr Tim Stevens Email: [log in to unmask]
Department of Biochemistry [log in to unmask]
University of Cambridge Phone: +44 1223 766022 (office)
80 Tennis Court Road +44 7816 338275 (mobile)
Old Addenbrooke's Site +44 1223 364613 (home)
Cambridge CB2 1GA WWWeb: http://www.bio.cam.ac.uk/~tjs23
United Kingdom http://www.pantonia.co.uk
-------------------------------------------------------------------------------
------ +NH3CH(CH(CH3)OH)C(O)NHCH(CH(CH3)CH2CH3)C(O)NHCH(CH2CH2SCH3)CO2- -------
-------------------------------------------------------------------------------
|