conMItion (version 0.2.1)

CMIBiCondimat2vecPermu: Permuted Normalized Conditional Mutual Information Between Matrix and Vector Given Two Conditions

Description

Computes the conditional mutual information (CMI) between a random vector sampled from a matrix and a vector, conditioned on two vectors, normalized by the mutual information of the vector with itself. The sampling is done multiple times to generate a distribution.

Usage

CMIBiCondimat2vecPermu(
  mat,
  vec,
  condi1,
  condi2,
  bin = 6,
  sp_order = 2,
  bulkIdx = 0,
  permutationTimes = 1000,
  seedNum = 99999999
)

Value

A numeric vector of normalized conditional mutual information (CMI) values for each permutation.

Arguments

mat

A numeric matrix. For example, each row represents a gene and each column represents a sample.

vec

A numeric vector, with length equal to the number of columns in `mat`.

condi1

A numeric condition vector, matching the number of columns in `mat`.

condi2

Another numeric condition vector, matching the number of columns in `mat`.

bin

An integer specifying the number of bins. Default is 6.

sp_order

An integer specifying the spline order. Must be less than `bin`. Default is 2.

bulkIdx

Index to divide the task when processing many permutations. Default is 0.

permutationTimes

Number of permutations for sampling. Default is 1000.

seedNum

Seed for random number generation. Default is 99999999.

Examples

mat <- matrix(rnorm(10000), nrow = 100, ncol = 100)
vec <- rnorm(100)
condi1 <- rnorm(100)
condi2 <- rnorm(100)
CMIBiCondimat2vecPermu(mat, vec, condi1, condi2)