triplex.search function identifies potential intramolecular
triplex-forming sequences in DNA.
triplex.search( dna, type = 0:7, min_score = 15, p_value = 0.05, min_len = 6, max_len = 25, min_loop = 3, max_loop = 10, seq_type = 'eukaryotic', score_table = 'default', group_table = 'default', lambda_par = 'default', lambda_apar = 'default', mu_par = 'default', mu_apar = 'default', rn_par = 'default', rn_apar = 'default', dtwist_pen = 'default', ins_pen = 'default', iso_pen = 'default', iso_bonus = 'default', mis_pen = 'default')DNAString object.
triplex.score.table output.
Before changing this option, please read triplex.score.table
help carefully.
triplex.group.table output.
Before changing this option, please read triplex.group.table
help carefully.
TriplexViews object based on
XStringViews class.
The triplex.search function identifies potential intramolecular
triplex-forming sequences in DNA sequence represented as
a DNAString object.
Based on triplex position (forward or reverse strand) and its third strand
orientation, up to 8 types of triplexes are distinguished by the function (see
the following figure). By default, the function detects all 8 types, however
this behavior can be changed by setting the type parameter to any value
or a subset of values in the range 0 to 7.
types.pngFigure 1: Triplex types
Detected triplexes are returned as instances of the
TriplexViews class,
which represents the basic container for storing a set of views on the same
input sequence similarly to the XStringViews object (in fact
TriplexViews only extends the XStringViews class
with a number of displayed columns). Each triplex view is defined by start
and end locations, width, score, P-value, number of insertions, type, strand, loop
start and loop end. Please note, that the strand orientation depends on
triplex type only. The triplex.search function assumes that the input
DNA sequence represents the forward strand.
Basic requirements for the shape or length of detected triplexes can be
defined using four parameters: min_len, max_len,
min_loop and max_loop. While min_len and max_len
specify the length of the triplex stem composed of individual triplets,
min_loop and max_loop parameters define the range of lengths
for the unpaired loop at the top of the triplex. A graphical representation of
these parameters is shown in the following figure. Please note, these
parameters also impact the overall computation time. For longer triplexes,
larger space has to be explored and thus more computation time is consumed.
triplex.pngFigure 2: Triplex scheme
The quality of each triplex is represented by its score value. A higher score
value represents a higher-quality triplex. This quality is decreased by several
types of imperfections at the level of triplets, such as character mismatch,
insertion, deletion, isomorphic group change etc. Penalization constants for
these imperfections can be setup using the following parameters: mis_pen,
ins_pen, iso_pen, iso_bonus and
dtwist_pen. Detailed information about the scoring function and
penalization parameters can be found in (Lexa et al., 2011). It is highly
recommended to see (Lexa et al., 2011) prior to changing any penalization
parameters.
The triplex.search function can output a large list containing tens of
thousands of potential triplexes. The size of these results can be reduced
using two filtration mechanisms: (1) by specifying the minimal acceptable
score value using the min_score parameter or (2) by specifying the
maximum acceptable P-value of results using the p_value parameter. The
P-value represents the probability of occurrence of detected triplexes in
random sequence. By default, only triplexes with P-value equal or less than
0.05 are reported. Calculation of P-value depends on two extreme value
distribution parameters lambda and mi. By default, these
parameters are set up for searching in human genome sequences. It is highly
recommended to see (Lexa et al., 2011) prior to changing either of the
lambda and mi parameters.
TriplexViews,
triplex.score.table
triplex.group.table
triplex.diagram,
triplex.3D,
triplex.alignment
# GAA triplet repeats involved in Friedreichs's ataxia
seq <- DNAString("GAAGAAGAAGAAGAAGAAGAAGAAGAAGAA")
# Search specific triplex types (see details section)
triplex.search(seq, type=c(2,3), min_score=10, p_value=1)
# Search all triplex types
t <- triplex.search(seq, min_score=10, p_value=1)
# Sort triplexes by score
t[order(score(t), decreasing=TRUE)]
Run the code above in your browser using DataLab