|
Internet Electronic Journal of Molecular Design - IEJMD, ISSN 1538-6414, CODEN IEJMAT
| ABSTRACT - Internet Electron. J. Mol. Des. December 2006, Volume 5, Number 12, 570-584 |
Structural Characteristics of the Nucleotides Pairing in RNA: Principal Component Analysis
Huai Cao, Weixian Cheng, Taohong Li, Xulin Pan, and Ciquan Liu
Internet Electron. J. Mol. Des. 2006, 5, 570-584
|
Abstract:
RNA secondary structure motifs are important to the biological
function of an RNA molecule. Close attention has been paid to the
pairing state of a nucleotide presented in a variety of motifs in many
reported studies of RNA structure and folding. More and more work
was crowed on identifying, classifying, and discovering motifs
because it is necessary that the interactions between secondary
structure motifs facilitate the higher order structure to come into being.
However, the paired–unpaired state for a nucleotide should be
alterable while a given functional higher order structure is resulted
from the secondary structure motifs. Studying the conformational
factors of affecting the state change of a motif and mastering the
factors to arrange the type of a motif is undoubtedly a key problem to
understand, find out, and further develop the biological activity of
RNA in the gene expression. In this article we collected the structural
data of 1320 nucleotides of 45 RNA molecules in the various motifs
from the databases, and analyzed the structural characteristics of their
torsion angles by using principal component analysis. We adopted
three grouping methods to analyze the 45 RNA molecules: in original
motif, in molecular type, and in new motif. The variables representing
conformation feature are six backbone torsion angles
α, β, γ, δ, ε, ζ and χ
between the sugar and the base, and the distance between the two
phosphorus atoms P. A pre–treatment of the data, the equal weight and
mean–centering for each variable, was used. The biplot representation,
in which the scores on the first two PCs together with the loadings of
the original variables are depicted in the same plot, was a proper
illustration for the PCA results. In the PC1–PC2 biplots, the
conformation feature of the nucleotides didn't display over–particular
with the variety of motifs, or the difference of molecules. Only the
three of the eight parameters, i.e.,
α, γ, and ζ, had the most loadings
and a negative correlation between α and γ. By comparison, the rest
parameters represented a little of contribution to PC1 and PC2.
Considering the scores on PC1, we grouped the nucleotides into the
two clusters: one gathering in the angle between the vectors α and γ,
and another dispersing out the angle. Significantly, the former were the
paired nucleotides, and the latter unpaired. All results revealed that the
conformational factors, α and ζ, especially α, play an irreplaceable role
in determining the pairing tendency of a nucleotide. Nucleotides in
RNAs may fall into two states according to the values of their torsion
angles, α and ζ, or only α, paired and unpaired. In comparison with the
previous studies on the characteristics of RNA secondary structure
motifs, this work shows that, by dominating a nucleotide having a
range of α values, we may transform motifs to design desired RNA
molecules.
|