minet (version 3.30.0)

build.mim: Build Mutual Information Matrix

Description

build.mim takes the dataset as input and computes the mutual information beetween all pair of variables according to the mutual inforamtion estimator estimator. The results are saved in the mutual information matrix (MIM), a square matrix whose (i,j) element is the mutual information between variables $Xi$ and $Xj$.

Usage

build.mim(dataset, estimator = "spearman", disc = "none", nbins = sqrt(NROW(dataset)))

Arguments

dataset

data.frame containing gene expression data or any dataset where columns contain variables/features and rows contain outcomes/samples.

estimator

The name of the entropy estimator to be used. The package can use the four mutual information estimators implemented in the package "infotheo": "mi.empirical", "mi.mm", "mi.shrink", "mi.sg" and three estimators based on correlation: "pearson","spearman","kendall"(default:"spearman") - see details.

disc

The name of the discretization method to be used with one of the discrete estimators: "none", "equalfreq", "equalwidth" or "globalequalwidth" (default : "none") - see infotheo package.

nbins

Integer specifying the number of bins to be used for the discretization if disc is different from "none". By default the number of bins is set to $sqrt(m)$ where m is the number of samples.

Value

build.mim returns the mutual information matrix.

Details

• "mi.empirical" : This estimator computes the entropy of the empirical probability distribution.
• "mi.mm" : This is the Miller-Madow asymptotic bias corrected empirical estimator.
• "mi.shrink" : This is a shrinkage estimate of the entropy of a Dirichlet probability distribution.
• "mi.sg" : This is the Schurmann-Grassberger estimate of the entropy of a Dirichlet probability distribution.
• "pearson" : This computes mutual information for normally distributed variable.
• "spearman" : This computes mutual information for normally distributed variable using Spearman's correlation instead of Pearson's correlation.
• "kendall" : This computes mutual information for normally distributed variable using Kendall's correlation instead of Pearson's correlation.

References

Patrick E. Meyer, Frederic Lafitte, and Gianluca Bontempi. minet: A R/Bioconductor Package for Inferring Large Transcriptional Networks Using Mutual Information. BMC Bioinformatics, Vol 9, 2008. J. Beirlant, E. J. Dudewica, L. Gyofi, and E. van der Meulen. Nonparametric entropy estimation : An overview. Journal of Statistics, 1997.

Jean Hausser. Improving entropy estimation and the inference of genetic regulatory networks. Master thesis of the National Institute of Applied Sciences of Lyon, 2006.

See Also

clr, aracne, mrnet, mrnetb

Examples

  data(syn.data)
  mim <- build.mim(syn.data,estimator="spearman")