getFeatures: getFeatures

In classSNitch: Classifies RNA Structure Change in Chemical Mapping Data

Description

A function to get the features for describing RNA structure change. These features can be used in classification of RNA structure change.

Usage

getFeatures(sample, base=NULL, margin=1,  norm=T, noise=T, mean=1.5, trim=0, high=NULL,
   tol=0.1, outfile=NULL, append=F)

Arguments

sample	A numeric matrix containing values to be compared (e.g. a set of mutant SHAPE traces).
base	An optional numeric vector containing the values to which the samples are to be compared (e.g. a wild type SHAPE trace). Default is the first trace in each file.
margin	An optional number indicating if the samples are organized by rows or columns, where 1 indicates rows and 2 indicates columns. Default is 1.
norm	An optional boolean to normalize the sample. Default is TRUE.
noise	An optional boolean to reduce noise in the sample. Default is TRUE.
mean	An optional number setting the mean SHAPE value for normalization. Default is 1.5.
trim	An optional number indicating the number of nucleotides to be trimed from the ends. Default is 0.
high	An optional number indicating the reactivity above which reactivities are considered high. Default is third quartile of the sample in each file.
tol	An optional number indicating the tolerance for the change. Default is 0.1.
outfile	An optional string indicating the name of the output file. The output file will consist of feature columns. Default will not output a file.
append	An optional boolean to append the file if an outfile is given. Default is FALSE.

Details

This function calculates the pattern correlation coefficient, dynamic time warping, contiguousness of change, magnitude correlation coefficient, change of variance, experimental structural disruption coefficient, change range and L2 norm. These are features used to describe structure change in SHAPE traces.

Value

"outmat"

A seven column numeric matrix for pattern correlation coefficient, dynamic time warping, contiguousness of change, magnitude correlation coefficient, change of variance, experimental structural disruption coefficient, change range, and L2 norm

"outfile"

An optional output file for the matrix.

Author(s)

Chanin Tolson

See Also

normalize reduceNoise getPatternCC getTimeWarping getContiguous getMagCC getESDC getChangeRange getL2norm

Examples

#input files
data("shape_ex")
#get features
params = getFeatures(shape_ex, trim=5, outfile="out.txt")

classSNitch documentation built on May 2, 2019, 5:51 p.m.