Definitions and Nomenclature of Parameters used to define Nucleic Acid
In 1989, a group of some sixteen prominent crystallographers, drawn from
many of the best-known research groups with an interest in oligonucleotide
structural studies, met in Cambridge to seek "...the creation of...routines
for the analysis and description of polynucleotide structure,
especially, but not exclusively the DNA double helix." (79,80,81)
In Figure 2 (79, page 788) are defined the
"...local reference axes (x,y,z) at the first two base pairs of an n
base-pair double helix."
On page 789 we read
"With these conventions, the signs of helical twist Omega, propeller twist
omega, displacement dx, and inclination eta, for three families of DNA
double helix typically are as follows:..."
The decisions taken in Cambridge certainly influenced some researchers at
least (82, page 657):
"The choice adopted in Cambridge concerning the axis system describing the
double helix leads us to modify the base fixed axis system (of CURVES) which
serves to locate any given nucleotide in space."
Understood literally, it seems evident that the conference participants in
Cambridge undertook to conduct future crystallographic studies of
oligonucleotides, at least, as they had perhaps been doing already anyway,
in such a way that there could be only one possible allowed structural
outcome, namely, the double helix.
It would not matter whether heavy atoms were used, or MAD, or synchrotron
sources, or the use of allegedly ab inititio methods, or whatever; since
only one structural outcome would be possible.
Many algorithms have been used to determine oligonucleotide structure, such
as, CORELS, DBLHLX, HELIX, NEWHELIX, NUCLSQ, SHELX93 and ULTIMA, for
example. Doubtless it would be unfair merely to assume that all these
programs, or the context of their use, incorporated directly or indirectly,
explicitly or implicitly, an a priori acceptance of the DNA double helix,
but, unfortunately, explicit details of the working of the algorithms, and
context of their use, do not seem to be widely and publicly available.
The experimental, published results identified in the Puzzles, and more
extensively elsewhere (42,64), pose a severe and so far unanswered test of
the double-helical paradigm.
It would seem appropriate therefore, in view of the Cambridge meeting (79 -
81) which appears to limit crystallographic outcomes solely to one duplex
model, and because of the continuing great importance of work with duplex
DNA in many fields of biochemistry, biophysics and molecular biology, for
researchers in these fields to consider asking their professional societies
to reflect upon the merits of recommending a moratorium on all future
citations of the work and results of studies in oligonucleotide
crystallography until certain, satisfactory clarifications have been
forthcoming in the literature.
These could include statements, supported by publication of the algorithms
in an accessible format, defining:
a) the precise philosophy underpinning the program and any relationship
it may have to the double helical paradigm,
b) the precise structural origin of any library data used by the
c) the precise character and purpose of any manual intervention by the
investigators themselves during refinement,
d) the precise origin of any data on models used in a study to help
determine the structure of a new oligonucleotide, and
e) the ready availability of all the structure factor files, many of
which are not accessible via the Brookhaven National Protein Data Bank
Until these conditions have been fully met, and the problem of the many
outstanding unexplained results reported in the literature, and reviewed
elsewhere (42, 64), has been fully addressed, the professional societies may
elect to consider that the oligonucleotide structures held at PDB and
elsewhere are insecure.
The current situation reflects that described by Francis Bacon in 1620:
"The human intellect, in those things which have once pleased it...brings
everything else to support and agree with it; and though the weight and
number of contradictory instances be superior, still either overlooks, or
despises, or gets rid of them...that the authority of these previous
conclusions may remain inviolate."
79 Definitions and Nomenclature of Nucleic Acid Structure Parameters;
R.E. Dickerson et al.; J Mol Biol Vol 205 (1989) 787 - 791
80 Definitions and nomenclature of nucleic acid structure components;
R.E. Dickerson et al.; Nucleic Acids Res Vol 17 (1989) 1797 - 1803
81 Definitions and Nomenclature of Nucleic Acid Structure Parameters;
Reported by R.E. Dickerson; J Biomol Struct & Dyn Vol 6 (1989) 627 - 634
82 Defining the Structure of Irregular Nucleic Acids: Conventions and
Principles; R. Lavery & H. Sklenar; J Biomol Struct & Dyn Vol 6 (1989) 655 -