findDistinguishingPCs: Find distinguishing PCs

In Carldeboer/BrockmanR: K-mer-based analysis of chromatin state.

Description

Identifies which PCs distinguish a feature of interest. If the feature of interest contains two classes, performs a single rank sum test for each PC. If the feature of interest has >2 classes, performs a rank sum test to see if the PC separates the class from all others. In either case, returns a data frame containing the PC, class, AUROC, and rank sum P-value for how well the PC identifies the class.

Usage

findDistinguishingPCs(rotatedData, classLabels)

Arguments

rotatedData	The samples' projections on PC space. Include here as many PCs as you want to inspect
classLabels	Either a data frame (first column containing sample labels, second containing classes), or a list where the names of the list are the sample labels and the entries in the list are the classes.

Value

Returns a data.frame containing the following columns: PC (which PC was used for each comparison), AUROC (the area under the ROC curve for how well each PC distinguishes each class), P (rank sum p-value for how well the PC distinguishes the class), class (the class).

Examples

kmerMat = inputKMerFreqs(sprintf("kMerFiles/%s.freq.gz",sampleDesc$id), IDs = sampleDesc$id)
myPCA = doKMerPCA(kmerMat, nPCs = "jackstraw")

Carldeboer/BrockmanR documentation built on May 31, 2019, 2:19 p.m.