I am surprised that the fuss about the structure 3J3Q
the pdb documentation states
"For all structures deposited as complex assemblies, we will archive
symmetry information as appropriate in _pdbx_point_symmetry or
_pdbx_helical_symmetry records."
For the cone shaped viral capsid discussed in 1999:
Assembly and Analysis of Conical Models for the HIV-1 Core
Science 1 January 1999: Vol. 283 no. 5398 pp. 80-83
DOI: 10.1126/science.283.5398.80
Barbie K. Ganser, Su Li, Victor Y. Klishko, John T. Finch, Wesley I.
Sundquist
and the symmetry properties explained in
Elasticity Theory and Shape Transitions of Viral Shells
T. T. Nguyen, R. F. Bruinsma, W. M. Gelbart
Phys. Rev. E 72, 051923 (2005)
DOI:10.1103/PhysRevE.72.051923
this paper describes the cone shape and the symmetry, i.e.
"In contrast, an example where continuum theory really is expected to
be applicable are core particles of the HIV-1 virus. The immature HIV
capsid is spherical, though not icosahedral, with diameters in the
range of 120 to 260 nm39. After cleavage of the Gag capsid protein
into CA ("capsid") and NC ("nucleocapsid") proteins – plus a matrix
protein – the core reforms into a conical shell with a size of about
100 nm (majority case) plus a smaller fraction of tubular particles.
The FvK number of the (mature) cone is of the order of 20,000, well
above the buckling threshold. According to our results, conical
capsids with FvK numbers in this range do constitute a well-defined
local minimum of the elastic energy with m/n ratios between 1.5 and 2
and spontaneous curvatures about twice the critical value (see
Fig.15)."
These 2 papers on symmetry of cones must be known to all concerned
isnt it possible to use this symmetry as NCS in the structure just
released that causes the fuss about large structures?
Simply cat the split*.pdb and rasmol clearly shows the symmetry of the
cone from the apex radiating out and recombining back to the other
end, of note is that rasmol colours chains at a TER record and the
colours are based on the order in the file so the split of this
molecule is very random.
Surely the authors and the PDB staff should have known about the
symmetry description or is everyone
stuck on the idea of Viral Capsid T-numbers which only cover basically
spherical shapes? At the resolution (actually not stated on the EMDB page)
http://www.ebi.ac.uk/pdbe/entry/EMD-5639/visualization
and using the volume viewer, it is unlikely that the authors could
have built the chains without the use of the cone symmetry and
at the map details shown there could not be any detectable break in the
use of symmetry. (The title page figure shown is
"simulated density of all-atom HIV-1 capsid model (3J3Y)
overlaid with a slice of HIV-1 core tomogram." so appears
not an experimental observation.
Actually just what is the difference between 3j3q and 3j3y
as there is no obvious difference on display and the mmcif titles
are no help.
_struct.entry_id 3J3Y
_struct.title 'Atomic-level structure of the entire HIV-1 capsid (186
hexamers
+ 12 pentamers)'
_struct.entry_id 3J3Q
_struct.title 'Atomic-level structure of the entire HIV-1 capsid'
And why is what was a part of the old pqs o/p for a virus showing
the 5fold and the 23fold being issued as yet another PDB idcode
as PDB 3J34 unspecified 'Hexameric unit obtained by MDFF and cryo-EM'
- which shows a core fragment, is this a theoretical model ?
The structure would have made less fuss if issued with BIOMOLECULE
matrices and been generated in full by software.
If the author can over ride PDB stated processing policy then what is
the use of the documentation? under
"Processing Procedures Version 2.6 as of December 2012"
on http://www.wwpdb.org/docs.html#format
as it is obviously not in use.
