PeakSegDP: PeakSegDP

In PeakSegDP: Dynamic Programming Algorithm for Peak Detection in ChIP-Seq Data

Description

Compute the PeakSeg model on a data.frame of compressed sequence reads.

Usage

PeakSegDP(compressed, maxPeaks)

Arguments

compressed	data.frame with columns chromStart, chromEnd, count.
maxPeaks	maximum number of peaks to consider.

Author(s)

Toby Dylan Hocking, Guillem Rigaill

Examples

library(PeakSegDP)
data(chr11ChIPseq, envir=environment())
one <- subset(chr11ChIPseq$coverage, sample.id=="McGill0002")[10000:12000,]
fit <- PeakSegDP(one, 3L)

if(interactive() && require(ggplot2)){

  ggplot()+
    geom_step(aes(chromStart/1e3, count), data=one)+
    geom_segment(aes(chromStart/1e3, mean,
                     xend=chromEnd/1e3, yend=mean),
                 data=fit$segments, color="green")+
    geom_segment(aes(chromStart/1e3, 0,
                     xend=chromEnd/1e3, yend=0),
                 data=subset(fit$segments, status=="peak"),
                 size=3, color="deepskyblue")+
    theme_bw()+
    theme(panel.margin=grid::unit(0, "cm"))+
    facet_grid(peaks ~ ., scales="free", labeller=function(df){
      s <- ifelse(df$peaks==1, "", "s")
      df$peaks <- paste0(df$peaks, " peak", s)
      df
    })

}

PeakSegDP documentation built on May 1, 2019, 7:55 p.m.