MAMA (version 2.2.1)

RandomScore: Function to do compute random and subsampled similarity score

Description

Function computes random and subsampled similarity score in order to select appropriate tunning parameter alpha for weights in preliminary simmilarity score

Usage

RandomScore(data, varname, B, alpha, test, which = c("random", "empirical", "subsample"), two.sided = TRUE)

Arguments

data

MetaArray object

varname

String indicating which column of clinical data should be used as class labels, same on all data set

B

Number of permutation

alpha

Vector of alphas considered (can be returned by computeAlpha) or selected manually

test

"FCH" for fold change (function fold.change) or "T" for T-test (function meta.test)

which

"random" for random score (permutation of class labels), "subsample" for subsampled score, "empirical" for empirical confidence intervals of overalaping genes, vector of several is also possible

two.sided

if TRUE both top and bottom of the ordering considered, if FALSE only top positions are considered

Value

A list

random

Random similarity score

empirical

Empirical confidence intervals for overlaping genes

subsample

Subsampled score