getPromoterClass-methods: Determining the CpG promoter class and the average CpG...
In compEpiTools: Tools for computational epigenomics

Description Methods Examples

Determining the CpG promoter class (Low, Intermediate or High CpG Content: lowCG, intCG or highCG, respectively) and the average CpG content for each entry of a transcriptDB.

To be used in this form:

getPromoterClass(txdb, Nproc=1, org, upstream=1000, downstream=0)

where:

txdb:An object of class TxDb
Nproc:numeric; the number of processors to be used; one chr is run for each processor
org:an object of class BSgenome
upstream:numeric; number of bp upstream transcription start sites defining upstream limit of promoters
downstream:numeric; number of bp downstream transcription start sites defining downstream limit of promoters

According to Weber M et al, Nature Genet 2007: the CpG content is determined as (W*CpG)/(C*G), where W is the window size and CpG, C and G are the number of CpG, C and G occurrences, respectively. Here W is set to 500bp. The CplusG content is (C*G)/W. The promoter CpG class is determined sliding a 500bp window in 1Kb upstream regions, with step of 5 bp. If the maximum CpG content for a given promoter is < 0.48, the promoter is assigned a lowCG. If the maximum CpG content is > 0.75 and CplusG>0.55, the promoter is assigned a highCG. The remaining promoters are assigned a intCG. The average CpG ratio is the average of the CpG ratios for all the windows.

A GRanges object decorated with the promoterClass and promoterCpG data is returned.

require(BSgenome.Mmusculus.UCSC.mm9)
require(TxDb.Mmusculus.UCSC.mm9.knownGene)
txdb <- TxDb.Mmusculus.UCSC.mm9.knownGene
isActiveSeq(txdb) <- c(rep(FALSE,20), TRUE, rep(FALSE, 14))
allpromoter <- getPromoterClass(txdb, Nproc=1, org=Mmusculus)
restoreSeqlevels(txdb)