corMatToAUC: Evaluate meanAUC/meanAUPR/medianAUC/medianAUPR of Recovering...

In wgmao/DataRemix: DataRemix: a universal data transformation for optimal inference from gene expression datasets

Description

Given the task of recovering known pathways/network, calculate the average AUC value and average AUPR value based on gene correlation matrix and the gold stanard, e.g., the known pathway matrix.

Usage

corMatToAUC(data, GS, objective = "mean.AUC")

Arguments

data	Correlation matrix with the dimension gene-by-gene
GS	GS stands for gold standard which is a binary matrix with the dimension gene-by-pathway. The number of rows should match that of data and the genes should be in the same order. This pathway matrix can be obtained from pathway databased, e.g., MSigDB
objective	"mean.AUC" (default option): Output is c(meanAUPR, meanAUC) and the objective is meanAUC (average AUC across all pathways) "mean.AUPR"" : Output is c(meanAUC, meanAUPR) and the objective is meanAUPR "median.AUC" : Output is c(medianAUPR, medianAUC) and the objective is medianAUC (median AUC across all pathways) "median.AUPR" : Output is c(medianAUC, medianAUPR) and the objective is medianAUPR

Value

A vector c(meanAUPR, meanAUC). By default, the biological objective is the average AUC acorss all pathways. The output will change based on objective

wgmao/DataRemix documentation built on Aug. 6, 2020, 4:49 p.m.