imprtance of knowing 3d structure

[Mr. G. Morley]

Well from personal experience knowing the tertiary structure of
the protein I am working on would be very handy...
As more and more genes are being found and sequenced the emphasis
is moving from "where are they" to "what do they do and how do
they do it?"
Currently to study how proteins function one of the most common 
methods used is mutagenesis. This involves mutating the DNA
encoding the receptor (either randomly or at a specific
codon) and examining the effects this mutation has had on
the normal function of the gene In Vivo (after expression in
cancer cell lines) or In Vitro (effects of kinase activity for
eg:). This approach is rather like trying to find out how
an engine works by taking out components of the engine one by
one whilst wearing a blindflod and then seeing what the effect 
this has on trying to start the car....
With a perfect model of a previously uncrystalised protein
you would probably still have to "replace car parts" to find out
the function of each domain or which areas (eg: ser, tyrosines, 
threonines) are binding sites for downstream signals, but at least you
could remove the blindfold. You would at least have some idea
of what amino acids interact with each other - and hence what effect
the replacement of a valine for an aspartate might have.
Another aspect of the tertiary structure would be to show what
domains are exposed and which are "buried" within the protein...
this might be important as a starting place to examine proteins which
may bind to the one understudy, or designing drugs to inhibit it.
 I could gibber on a bit longer on this (we would not have to
do x-ray crystalography anymore - which would save a lot of time and money)
but I hope this has given you my impression at least (and I am
by no means an authority on this!).
I would be happy to hear anyone elses views on this topic..
regards,

		Gary Morley
		gmorley at rpms.ac.uk