test2grp {wrProteo}

t-test each line of 2 groups of data

Description

test2grp performs t-test on two groups of data using limma, this is a custom implementation of moderTest2grp for proteomics. The final obkect also includes the results without moderation by limma (eg BH-FDR in $nonMod.BH). Furthermore, there is an option to make use of package ROTS (note, this will increase the time of computatins considerably).

Usage

test2grp(
  dat,
  questNo,
  useCol = NULL,
  grp = NULL,
  annot = NULL,
  ROTSn = 0,
  silent = FALSE,
  debug = FALSE,
  callFrom = NULL
)

Arguments

dat

(matrix or data.frame) main data (may contain NAs)

questNo

(integer) specify here which question, ie comparison should be adressed

useCol

(integer or character)

grp

(character or factor)

annot

(matrix or data.frame)

ROTSn

(integer) number of iterations ROTS runs (stabilization of reseults may be seen with >300)

silent

(logical) suppress messages

debug

(logical) display additional messages for debugging

callFrom

(character) allow easier tracking of message(s) produced

Value

This function returns a limma-type S3 object of class 'MArrayLM' (which can be accessed like a list); multiple testing correction types or modified testing by ROTS may get included ('p.value','FDR','BY','lfdr' or 'ROTS.BH')

See Also

moderTest2grp, pVal2lfdr, t.test, ROTS from the Bioconductor package ROTS

Examples

set.seed(2018);  datT8 <- matrix(round(rnorm(800)+3,1), nc=8, dimnames=list(paste(
  "li",1:100,sep=""), paste(rep(LETTERS[1:3],c(3,3,2)),letters[18:25],sep="")))
datT8[3:6,1:2] <- datT8[3:6,1:2] +3   # augment lines 3:6 (c-f) 
datT8[5:8,5:6] <- datT8[5:8,5:6] +3   # augment lines 5:8 (e-h) 
grp8 <- gl(3,3,labels=LETTERS[1:3],length=8)
datL <- list(data=datT8, filt= wrMisc::presenceFilt(datT8,grp=grp8,maxGrpM=1,ratMa=0.8))
testAvB0 <- wrMisc::moderTest2grp(datT8[,1:6], gl(2,3))
testAvB <- test2grp(datL, questNo=1)