mQTL.NMR (version 1.6.0)

pre_mQTL: Statistical Recoupling of variables for mQTL analysis

Description

Makes use of SRV to preprocess metabolomic data for dimensionality reduction by statistical recoupling of variables

Usage

pre_mQTL(infile, outfile, RedMet="SRV", met="sum", corrT = 0.9, BinWidth=0.01)

Arguments

infile

metabolomic datafile in csvs format

outfile

reduced metabolomic datafile in csvs format

met

a charcater specifying the used statistical summary

RedMet

a charcater indicating the used dimensionality reduction method: Redmet="SRV" for statistical recoupling of variables and Redmet="bin" to apply the bining approach

corrT

a numerical parameter indicating correlation threshold

BinWidth

a numerical parameter indicating the bining width

Value

variables are associated into a series of clusters (or bins). This function provides in output the paramaters of the clusters (min and max borders, mean,...)

Details

mQTL-NMR package implements two dimensionality reduction methods. The first one concerns the SRV algorithm which forms clusters of variables using a measure of a local spectral dependency. The second one concerns the classical bining method which divides the spectra into evenly spaced windows (bins) whose width commonly ranges between 0.001 and 0.05 ppm.

References

- Blaise,B. et al (2009) Statistical recoupling prior to significance testing in nuclear magnetic resonance based metabonomics, Anal. Chem., 81(15), 6242-6251. - S praul, M.; Neidig, P.; Klauck, U.; Kessler, P.; Holmes, E.; Nicholson, J. K.; Sweatman, B.C.; Salman, S.R.; Farrant, R.D.; Rahr, E.; et al. J.Pharm. Biomed. Anal. 1994, 12, 1215-1225.

See Also

SRV,post_mQTL

Examples

# Download data files

load_datafiles()

# Format data

format_mQTL(phenofile,genofile,physiodat,cleandat,cleangen)

# Constant Sum normlisation
nmeth<-'CS'
normalise_mQTL(cleandat,CSnorm,nmeth)

# Alignment
align_mQTL(CSnorm,aligdat)

# Dimensionality reduction
met="rectangle" # choose the statistical summarizing measure ("max","sum","trapez",...)
RedMet="SRV" # reduction method ("SRV" or "bin")

pre_mQTL(aligdat, reducedF, RedMet="SRV",met, corrT=0.9)