Rasmus Fogh wrote:
> Dear All,
>
> Unlike most other programs we do not have (the equivalent of) a single
> resonance for e.g. Lys HE*. I still think ours is the correct description.
> Systems that are symmetric or in fast exchange, like CH3 or flipping
> phenylalanine are degenerate, and get one resonance corresponding to
> several protons.  Prochirals are not degenerate, they just happen to have
> the same shift in one particular spectrum. I'd say that representing them
> as two resonances with identical shifts describes the situation perfectly.
>   
I would agree up to this point, this is indeed a situation that is 
different from methyls and aromatics. So from a theoretical point of 
view you are absolutely right.
> To be sure it does mean that you have to remember to assign all peaks as
> ambiguous 'HE2 or HE3', or errors will creep in.
>   
This however is a problem. Because if you forget to assign both of your 
Hb2/Hb3 resonances to a given NOE peak, you will not only generate 
stereospecific information where you don't actually have that 
information but it will also have the wrong intensity information 
because the underlying peak is most likely the overlay of two peaks.  
> I preferred HE2/HE3 instead of HEa/HEb because it contains more
> information. If the shifts really are identical you *do* know for certain
> the shift of both stereoassigned protons.
True, but to me this information seems to be irrelevant in most 
situations, _because_ the shifts are not resolved.   
>  If you do not like it, you still
> have the  option of assigning ambiguously to HEa/HEb, or to just HEa if
> you are not sure they are both there. I am more open to argument on this
> one, though.
>
> We also considered another alternative: to remove HE* (etc.) from the
> program completely. It does not correspond to the underlying information,
> being just a shorthand for something else, and people tend to get tripped
> up because it does not behave the way they expect.
>   
Because of the above issues, I personally would prefer if  unresolved 
prochirals were handled in the same way as degenerate ones. With a 
single resonance that is assigned to Hb*.  This would make working with 
them much more straightforward even if the educational aspect is lost. 
Personally I believe that Analysis should make our daily work easier, 
even if some clarity on underlying principles are lost. If you prefer, 
you could use Hb# for unresolved and Hb* for degenerate prochirals.

Just my 2 Cents,
Christoph

-- 
---------------------------------------------------------
Dr. Christoph Brockmann	
Division of Structural Studies
MRC Laboratory of Molecular Biology
Hills Road
Cambridge, CB2 2QH
phone: +44-1223-40-2261 or +44-1223-40-2042
e-mail: [log in to unmask]
---------------------------------------------------------