cosinePerm: Cosine Permuations

In PharmacoGx: Analysis of Large-Scale Pharmacogenomic Data

Description

Computes the cosine similarity and significance using permutation test. This function uses random numbers, to ensure reproducibility please call set.seed() before running the function.

Usage

cosinePerm(
  x,
  y,
  nperm = 1000,
  alternative = c("two.sided", "less", "greater"),
  include.perm = FALSE,
  nthread = 1
)

Arguments

x	factor is the factors for the first variable
y	factor is the factors for the second variable
nperm	integer is the number of permutations to compute the null distribution of MCC estimates
alternative	string indicates the alternative hypothesis and must be one of ‘'two.sided'’, ‘'greater'’ or ‘'less'’. You can specify just the initial letter. ‘'greater'’ corresponds to positive association, ‘'less'’ to negative association. Options are 'two.sided', 'less', or 'greater'
include.perm	boolean indicates whether the estimates for the null distribution should be returned. Default set to 'FALSE'
nthread	integer is the number of threads to be used to perform the permutations in parallel

Value

A list estimate of the cosine similarity, p-value and estimates after random permutations (null distribution) in include.perm is set to 'TRUE'

Examples

x <- factor(c(1,2,1,2,1))
y <- factor(c(2,2,1,1,1))
cosinePerm(x, y)

Example output

$estimate
[1] 0.9090909

$p.value
[1] 0.174

PharmacoGx documentation built on Feb. 28, 2021, 2 a.m.