kfold.Divergence: Leave-one-out cross-training for parametric divergence...

In MEET: MEET: Motif Elements Estimation Toolkit

Description

Given a training sequence set, the optimal value for parametric divergence has been estimated by means of leave-one-out cross-training from q-value set. For each q-value, the ROC curve has been calculated. From this results, the optimal q-value has been considered according to the area under convex surface maximum.

Usage

kfold.Divergence(iicc, TF)

Arguments

iicc	A set of inicial conditions for the MEET-package (mode, method, background, alignment, threshold, parameters, Transcriptionfactor, nummotif, lenmotif, sentit, position, missing, vector, gapopen, maxiters, gapextend)
TF	A set of nucleotide sequences

Details

This function integrates the Mutual information (Renyi Order equal 1) and parametric divergence (Renyi Order different 1). Moreover, it contains a set of function for the detection of transcription factor binding sites:correction.entropy.R, correction.redundancy.R, entropy.R entropy.max.R, entropy.corrected.R, probability.R, CalculRedundancy.R, diff.ructions.R, redundancy.R, missing.fun.R, ROC.R, detector_2nOrdre.R, pvalue.R.

Author(s)

Joan Maynou <[e-mail]>

References

J. Maynou, M. Vallverdu, F. Claria, J.J. Gallardo-Chacon, P. Caminal and A. Perera, Transcription Factor Binding Site Detection through Position Cross-Mutual Information variability analysis. 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

See Also

kfold.Entropy, kfold.MEME, kfold.MDscan, kfold.MATCH and kfold.PCA

Examples

data(iicc)
pathMEET<-system.file("exdata",package="MEET")
kfold.Divergence(iicc,TF=paste(pathMEET,"AP1.fa",sep="/"))

MEET documentation built on May 2, 2019, 5:52 p.m.