MnM.selectDMR: Select significants of each comparation.

In methylMnM: detect different methylation level (DMR)

Description

The function is used to select significants of each comparation.

Usage

MnM.selectDMR(frames = NULL, up =1.45, down = 1/1.45, p.value.MM = 0.01,
 p.value.SAGE = 0.01,q.value = 0.01,cutoff="q-value",  quant= 0.6)
  

Arguments

frames	The input dataset and q-values of each bin.
up	The ratio of Medip1/Medip2 should be larger than "up" value if we call it significant.
down	The ratio of Medip1/Medip2 should be smaller than "down" value if we call it significant.
p.value.MM	The p-value of the bin which use MM test should be smaller than "p.value.MM" if we call it significant.
p.value.SAGE	The p-value of the bin which use SAGE test should be smaller than "p.value.SAGE" if we call it significant.
q.value	The q-value of the bin should be smaller than "q.value" if we call it significant.
cutoff	We should use p-value or q-value cutoff to detect DMRs (If cutoff="q-value", then we use q-value to detect DMRs, else we use p-value ).
quant	The rank of absolute value of (Medip1-Medip2) should be larger than "quant" value if we call it significant.

Value

The DMRs of the comparation.

Author(s)

Yan Zhou, Bo Zhang, Nan Lin, BaoXue Zhang and Ting Wang

Examples

  datafile<-system.file("extdata",  package = "methylMnM")
  filepath<-datafile[1]
  file1<-paste(filepath,"/all_CpGsite_chr18.txt",sep="")
  CpGsite<-read.table(file1, header=FALSE,skip=0, nrows=200, as.is=TRUE) 
  winbin<-CpGsite[1:100,1:4]
  winbin[,2]<-seq(0,49500,500)
  winbin[,3]<-winbin[,2]+500
  count<-matrix(rpois(600, lambda=5), nrow=100 )
  count[,6]<-count[,5]
  pvalue<-runif(100, min=0, max=1)
  ts<-rnorm(100, mean=0, sd=1) 
  cpgpq<-cbind(winbin[,1:3],count,pvalue,ts)
  colnames(cpgpq)=c("chr", "chrSt","chrEnd","Medip1","Medip2","MRE1",
  "MRE2","cg","mrecg","pvalue",'Ts')
  f<-MnM.selectDMR(frames=cpgpq, p.value.MM = 0.1, 
  p.value.SAGE = 0.1,cutoff="p-value")

Example output

Loading required package: edgeR
Loading required package: limma
Loading required package: statmod
[1] Total number of frames: 100
[1] Remaining number of frames with MRE-CpG=0 and p.value<=0.1: 0
[1] Remaining number of frames with MRE-CpG!=0 and p.value<=0.1: 12
[1] Remaining number of frames with D_value>=2& CpG!=0: 8
[1] Remaining number of frames with ratio<=0.689655172413793 or ratio>=1.45: 6
[1] [Note: There are 0 frames associated to a p-value==0.]
[1] Remaining number of frames called DMR:6

methylMnM documentation built on Nov. 8, 2020, 6:47 p.m.