rowcorMatVec: Calculate correlation between rows of a matrix and a vector

Description Usage Arguments Value

View source: R/TestAndCorr_p.R

Description

Calculate correlation between rows of a matrix and a vector

Usage

1
2
3
4
5
rowcorMatVec(mat, vec, meta = list(tag = "correlation", row = "expression",
  response = "response"), method = "spearman", minN = 5, usercorr = TRUE,
  addData = TRUE, symbol = NULL, sepSymProbe = ", ", FDR = NULL,
  pcutoff = 0.05, alphas = c(0.01, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5),
  addBum = NULL)

Arguments

mat

a matrix, i.e. gene expression

vec

a vector of phenotype, continuous variable

meta

meta information, a list. This can specify what type of analysis (i.e. RPPA correlated with EMT score); row, which represents what row variables are (rows of mat); response represents what does vec mean

method

method passed to correlation calculation. It can be spearman or pearson

minN

minimum sample size for correlation calculation; If not achieved, correlation is returned as NA

usercorr

whether to use rcorr package for speeded calculation (using Fortran)

symbol

in case this is probe level data, user can specify a vector of symbols matching

sepSymProbe

separator for symbol-probe. That is, symbol+sepSymProbe+probe(rowname) is the new rowname

FDR

FDR to create bum for the p values, passed to create_bum()

pcutoff

pcutoff passed to create_bum()

alphas

alphas passed to create_bum()

Value

a matrix with two columns: cor and pval


nickytong/GenAnalysis documentation built on July 20, 2019, 8:57 a.m.