Dear Brian,
You wrote:
>
> PS - could someone enlighten me on this question about experiment
> prototypes? It seems that experiment prototypes are ascribed that the
> level of experiments, but that there is no way of remapping the order of
> dimensions in individual instances of spectra. Is that right? Although
> for use with analysis, this should not be a problem (since you can always
> adjust how you see the spectra), there may be processing reasons, and
> certainly were good reasons in other software, as to why you might choose
> to reorder dimensions. If my interpretation of the model/software is
> right then you wouldn't be able to store differently ordered versions of
> the same experiment together, or if you did the prototype would be wrong
> for some. Appologies if I'm just being dumb.
>
The NmrExpPrototypes are linked to the Experiment. You can change the
mapping, but it is at the experiment level. I think that is right, what is
on the axes of *the experiment* does not depend on how you process it
later.
The mapping between the Experiment and the Spectra (DataSources) is set
explicitly when you create the Spectrum - it is the AbstractDataDim.expDim
link. In the model that is settable, though it has to be chosen at
creation time. Analysis currently assumes that the Spectra have the same
number of dimensions as the Experiment and that they appear in the same
order. This could clearly be changed, but it does require user input at
the time of Spectrum creation/entry. Question is how it should be changed,
and how urgent it is. In the meantime you could in theory create you own
Spectra by hand, on the command line. It would work provided you got it
right, but I know few people are likely to try.
Yours,
Rasmus
|