SRV: Statistical Recoupling of Variables

In lhedjazi/mQTL.NMR: Metabolomic Quantitative Trait Locus Mapping for 1H NMR data

Description

Base function for dimensionality reduction by statistical recoupling of variables

Usage

SRV(X, minsize, correl, clustf = median)

Arguments

X	matrix of metabolomic data
minsize	a numerical value defining the singlet size
correl	a numerical value defining the bucketting resolution
clustf	a numerical value defining the correlation threshold

Value

A list:

indicesdebf	a vector indicating the starting border of superclusters
indicesfinf	a vector indicating the ending border of superclusters
Xcluster	matrix of reduced data

Author(s)

Jean-Baptiste Cazier

References

Blaise,B. et al (2009) Statistical recoupling prior to significance testing in nuclear magnetic resonance based metabonomics, Anal. Chem., 81(15), 6242-6251.

See Also

pre_mQTL

Examples

# Load data files

    load_datafiles()

    Sp<-read.table(phenofile, as.is=TRUE, header=TRUE, sep='\t')

# Perform the SRV analysis to reduce the number of dimension of Spectra #data (Sp)

    corrT=0.9 # correlation threshold
    minsize=10 # singlet size
    met="rectangle" # summary measure

    SRV<-SRV(t(Sp), minsize, corrT,clustf=met)

lhedjazi/mQTL.NMR documentation built on May 31, 2019, 10:47 p.m.