Information Technology Reference
In-Depth Information
Figure 9.
Distribution of curvature around open reading frames in yeast
chromosome III. Curvature profiles of all fragments 200 bp outside and 50 bp
inside the open reading frame were averaged and the result is displayed in terms of
a positional preference for curved regions with respect to start and stop codons.
The black line represents the average curvature of yeast chromosome III.
Application examples
One of the obvious applications is to compare the distribution of curvature and other
parameters in genomic sequences. Figure 7 shows that bendability has a smooth,
symmetrical distribution in genomic DNA, similar to a bell shape. The distribution of
curvature is apparently non-symmetrical reminiscent of a gamma function which is often
found with randomly distributed variables whose value cannot be negative - curvature is
actually such a case. Another possibility is to analyse curved segments along the entire
genome. A circular plot is a convenient way to show such distributions even though the
graphic resolution is often a limiting factor. Another possibility is to analyse the vicinity of
annotated features in genomes, as shown in Figure 9. A comprehensive analysis of the
curvature of the
B. subtilis
genomic DNA revealed the percentage of curved motifs within
the genome and how many ORFs contain curved segments [43]. As reported in Figure 10,
less than 1% of the
B. subtilis
genome contains curved motifs with values above 14° per
helical turn. Using this as a cut-off, the majority of the curved DNA is found within the
ORFs while using 16°, 64% of the curved segments are within the intergenic regions
(Figure 10, inset), a tendency that continues as the cut-off is raised. In other words,
the majority of the most curved segments are concentrated in the intergenic regions.Figure
10 shows the number of ORFs with at least one curved motif. Only 6.2% of all the ORFs in
B. subtilis
shows a curvature with a cut-off value of 14°. These ORFs with at least one
curved motif, encode functionally unrelated proteins since their percentage distribution is
consistent with the distribution of the known proteins among the different classes,
following the standard functional classification reported by SubtiList (cellular processes
and cell envelope, intermediary metabolism, information pathways, other functions, [44]).
Therefore, only a small percentage of all curved motifs fall inside the coding regions,
leading to the hypothesis that a straight DNA is more efficiently transcribed. On the other
hand, it could be that intergenic regions have been selected with an intrinsic high curvature
to act as genomic signals. Indeed, it is known that, at least in lower eukaryotes such as
Saccharomyces cerevisiae
and
Leishmania major
, promoters and terminators are
constituted by flexible DNA stretches [45]. Several coding strand switching points are
present within the chromosomes of
L. major
[46]. For example in a region of 74 kb of
