On Sun, 2011-01-09 at 13:45 -0600, Kenneth A. Satyshur wrote:
> So lets include them in all refinements, regardless of resolution. At
> least we
> will have given the community a model with the most correct biological
> representation.
First of all, the fraction of structures in which the hydrogens can be
explicitly refined is very small, so the major effect of your
proposition will be the deposition of riding hydrogens. This has been
discussed plenty of times before, and I think what you are saying nicely
illustrates the dangers of this.
One very good argument for the deposition of riding hydrogens is that it
helps reproducing the exact R-values for the deposited model. While
this is undoubtedly useful for comparing refinement algorithms etc, most
uses of the PDB models revolve around model interpretation. Since
hydrogens are not under most circumstances (or at least should not be)
refined, their specific positions are simply projections from the
positions of heavier atoms. These may be quite accurate for something
like amide protons, but to deposit a set of three HZ* hydrogens for
every lysine sounds absurd.
There are other things that are often omitted from crystal structures,
such as disordered loops and side chains, water molecules in bulk
solvent regions and sometimes whole domains whose spatial position
cannot be determined from electron density with any certainty. Your
argument (as I understand it, so corrections are welcome) is that it is
misleading to present crystal structures without hydrogens because we
know they are there. But wouldn't it be misleading too to present
crystal structures with hydrogens at specific positions given that no
evidence is available to pinpoint their exact location?
A lot of this is related to potential danger of over-interpretation by
non-experts. It seems to me that models without hydrogens (and perhaps
without disordered side chains) are more likely to result in such person
seeking expert advice instead of directly deducing the protonation state
from 2.8A crystal structure.
Cheers,
Ed.
--
"I'd jump in myself, if I weren't so good at whistling."
Julian, King of Lemurs
|