wMICA.Run: wMICA: weighted Maximal Information Component Analysis

In ChristophRau/wMICA: weighted Maximal Information Component Analysis

wMICA: weighted Maximal Information Component Analysis

Description

wMICA.Run is the core function of the wMICA package and runs the entire package to generate weighted module memberships using the MINE algorithm and a modified weighted version of the ICMg algorithm.

Usage

wMICA.Run(input,output_prefix="MICA",n.cores=detectCores()-1,
threshold=.4,C=20,Seeds=c(),Weight=0,alpha=10,beta=.01,
B.num=10,B.size=10,S.num=10,S.size=10) 

Arguments

input	A NxM matrix where N are samples and M are genes
output_prefix	prefix to affix to outputted files
n.cores	Number of cores of computer to use when calculating MINE score. Defaults to one fewer than the total number of cores on the machine
threshold	lowest value edge to consider. A lower threshold (<=.3) is more accurate, but takes significantly longer
C	The Number of Modules
Seeds	a Kx2 matrix (where K is the number of genes to be seeded into a module) with the first column containing the gene number (as in L) and the second column being the desired module to be seeded into
Weight	the number of iterations to weight each seeded gene by
alpha	Hyperparameter describing the global distribution over components, larger alpha gives a more uniform distribution.
beta	Hyperparameter describing the component-wise distributions over nodes, larger beta gives a more uniform distribution.
B.num	Number of burnin rounds.*
B.size	Size of one burnin round.*
S.num	Number of sample rounds.*
S.size	Size of one sample round.*

Details

* One run involves three components. First, the Maximal Information score between each pair of genes is calculated. Then, Edges with a MIC score greater than the threshold are compiled and given to the module identification algorithm. Modules are determined using ICMg, whose details follow:

* One run consists of two parts, during burnin the sampler is expected to mix, after which the samples are taken. Information about convergence (convN and convL are estimates of convergence for link and node sampling, respectively) and component sizes are printed after each burnin/sample round. For example: B.num=8, B.size=100, S.num=20, S.size=10, runs 800 burnin iterations in 8 rounds and then takes 20 samples with an interval of 10 iterations.

*Finally, Module Memberships are extracted and all files (MIC array, Links, Memberships) are saved.

Value

Returns 3 output files to the working directory with prefix X:

X_MICARRAY.csv	N*N matrix of Maximal Information Scores
X_Links.csv	All included links used in wMICA
X_MODULEMEMBERSHIPS.csv	C*N matrix of results, where the rows are the modules and the columns are the genes

Author(s)

Christoph Rau

Examples

	## Load data and set parameters
	data(MICASample)
	Seeds=c()
  Weight=0
  C=20
  wMICA.Run(indata,Seeds=Seeds,Weight=Weight,C=C)

ChristophRau/wMICA documentation built on May 3, 2024, 3:21 a.m.