> Likewise, the argument that disordered atoms will inflate the B
> factors of the ordered atoms (like James Irving mentioned) is
valid
> the other way around too: an ordered C-beta will keep the B
factor
> for C-gamma at bay, and so on. If the B factor is meant to
reflect
> degree of disorder, then a mutual effect is completely
desirable.
> However, we all know that B factors are not perfect, either.
As has been mentioned in a previous email, the current B-factor
restraints on e.g. lysine side chains are far too restrictive. Since the
CA-CB bond is rotatable, the CG atom can be anywhere on a cone around
the CB, while the CB itself stays exactly in place. This effect is
clearly visible in many high-resolution structures where the CB is often
perfectly visible with no electron density whatsoever visible beyond
that. In my opinion, the B-factor restraints of the major refinement
packages should be adapted to make a distinction between rotable and
non-rotatable bonds. This makes physically far more sense than the
current distinction between main chain/side chain restraints and the
artifacts mentioned above would no longer occur.
Concerning depositing pdb files: if both the user and the software have
no clue about disordered side chains, any meaningfull interpretation of
the structure is impossible. In my view we should deposit the files such
that they best describe the physical reality present in the crystal and
not compromise the structures we deposit by concerns about ignorant
users and inadequate software. Instead we should educate users and
software developers to be aware of disorder and have the display program
clearly flag any potential problem areas.
Herman Schreuder
