parCorPatternComparison: Compare an input pattern against a set of patterns, excluding...

View source: R/parCorPatternComparison.R

parCorPatternComparisonR Documentation

Compare an input pattern against a set of patterns, excluding the predictive effect of a fixed pattern or set of patterns.

Description

Compare an input pattern against a set of patterns, excluding the predictive effect of a fixed pattern or set of patterns.

Usage

parCorPatternComparison(x, Y, Z, updateProgress = NULL)

Arguments

x

An N element input pattern specified as either a vector or a 1 x N matrix or data frame.

Y

An N element pattern specified as a vector for comparison with the input pattern x or a k x N matrix with k patterns for comparison with the input pattern x specified along the rows, with rownames set appropriately.

Z

An N element pattern specified as a vector or a k x N matrix of patterns specified along the rows. These are the patterns whose effect (with respect to a linear model) is to be excluded when comparing x with Y or each row entry of Y. Note that for the partial correlation to be value, the pattern(s) in Z should not overlap with those in x or Y.

updateProgress

A optional function to be invoked with each computed partial correlation to indicate progress.

Value

A data frame with pattern comparison results (ordered by PARCOR): NAME: Name of entry in Y being compared. PARCOR: Partial correlation between x and the entry in Y with respect to Z. PVAL: p-value.

Examples

x <- exprs(getAct(rcellminerData::drugData))["609699", ]
Y <- rcellminer::getAllFeatureData(rcellminerData::molData)[["exp"]][1:100, ]
Z <- rcellminer::getAllFeatureData(rcellminerData::molData)[["exp"]][c("SLFN11", "JAG1"), ]
results <- parCorPatternComparison(x, Y, Z)
Y <- rcellminer::getAllFeatureData(rcellminerData::molData)[["exp"]][1, , drop=TRUE]
Z <- rcellminer::getAllFeatureData(rcellminerData::molData)[["exp"]]["SLFN11", , drop=TRUE]
results <- parCorPatternComparison(x, Y, Z)


CBIIT/rcellminer documentation built on Aug. 8, 2024, 12:15 p.m.