gapsIntraPattern: `gapsIntraPattern` generates statistics for the similarity of gene expression vectors within a pattern

Description

gapsIntraPatternObtains correlation matrices for significantly expressed genes in each column of ${\bf{A}}$. Obtains the means of these matrices, then averages those to get a sense of how closely correlated genes in the patterns of CoGAPS are.

Usage

gapsIntraPattern(Amean, Asd, DMatrix, sdThreshold = 3)

Arguments

Amean

Sampled mean value of the amplitude matrix ${\bf{A}}$ from a run of CoGAPS.

Asd

Sampled standard deviation of the amplitude matrix ${\bf{A}}$ from a run of CoGAPS.

DMatrix

data matrix

sdThreshold

How many standard deviations a gene's sampled mean needs to be above 0 to be considered significantly expressed in a pattern.

Value

CorrelationMatrices: A list containing the correlation matrices between the significant genes in each column of ${\bf{A}}.$
CorrelationMatrixMeans: A list containing the means of the Correlation Matrices.
IntraPatternValue: The final value of the Intrapattern measure.

Details

This calculates a statistic for determining how tight a pattern is based on the genes associated with it and their distribution in the data.

Examples

Run this code

## Not run: 
# ## Load data
# data('SimpSim')
# 
# ## Run GAPS matrix decomposition
# nIter <- 10
# results <- gapsTestRun(SimpSim.D, SimpSim.S, nFactor=3,
#                    nEquil=nIter, nSample=nIter)
# 				   
# ## Execute the IntraPattern function
# IntraPatternStats <- gapsInterPattern(results$Amean, results$Asd, SimpSim.D, sdThreshold = 3)
# ## End(Not run)