Gerard Kleijwegt (gerard at rigel.bmc.uu.se) wrote:
: In article <3g8f7c$8hv at lyra.csx.cam.ac.uk>, smb18 at mole.bio.cam.ac.uk (Simon Brocklehurst (Bioc)) writes:
: May I take the opportunity to suggest a simple test to see if homology
: models are better than random ? Assume the structure of protein X
: is known, and that it is > 30 % homologous to protein Y. Now, if someone
: collects xray data of protein Y, a homology model of protein Y should
: at the very least give a clearly better solution in the Molecular
: Replacement than the structure of protein X. If not, the homology
: model does not contain more information than the structure of the
: 'undisturbed' protein X itself (and is therefore equally good as anyone's
: guess, where the xtallographer's guess would be to use a stripped
: version of protein X).
This is a lot of baloney !!, typical crystallographer's self-protecting
speech. The fact of the matter is that if your protein Y is a totally
new enzyme you can dramatically speed up your biochemical
characterization if you can model its structure.
Molecular modeling is a low resolution technique by comparision to
experimental methods, that is implicit in the system. However it can
provide information several orders of magnitude faster than crystallography.
To say that a model should provide a good solution to a molecular
replacement search is also nonsense. That is not what should be
expected from a model.
: - even if the fold can be predicted, this usually carries very little
: biologically-relevant information. The most interesting parts of a
: protein are often found in one or more of its loops. Moreover, at the
: end of the day, you'll only be interested in the conformation of a
: handful of side chains (the active site, a metal-binding site, a ligand-
: binding site, a protein-protein or protein-DNA interface), something
: which even with 80 % homology is often hard to do. And nature is always
: full of surprises.
Once again you confuse the subject, Nature has surprises even for the most
studied of structures (i.e. haemoglobin), what counts is the level of certanty
that you expect to obtain from your analysis. The whole bunch of methods that
we use today for protein modeling (from sequence alignments to structure analysis
by statistical force fields) provide us with good tools to get a good initial
approximation to a problem.
No modeller will never refuse to use x-ray data, and no crystallographer should
ignore the information that modeling techniques can provide. The RAS structure
would have been proved wrong by several modeling statistical methods had they been
used to analyze it.
Don't you think that part of the crystallographer's despise for modeling stems
from they anger at loosing the monopoly of structure determination ?.
Lluis
