Anreicherung bewerten

ChIP-seq mit Bioconductor in R

Peter Humburg

Statistician, Macquarie University

Reads erweitern

Daten laden:

reads <- readGAlignments(bam)
reads_gr <- granges(reads[[1]])

Durchschnittliche Fragmentlänge ermitteln:

frag_length <- fragmentlength(qc_report)["GSM1598218"]

Reads erweitern und Coverage berechnen:

reads_ext <- resize(reads_gr, width=frag_length)
cover_ext <- coverage(reads_ext)

Coverage für Peaks

Erzeuge 200-bp-Bins entlang des Genoms.

bins <- tileGenome(seqinfo(reads), tilewidth=200, 
                   cut.last.tile.in.chrom=TRUE)

Finde alle Bins, die Peaks überlappen.

peak_bins_overlap <- findOverlaps(bins, peaks)
peak_bins <- bins[from(peak_bins_overlap), ]

Zähle die Reads, die jedes Peak-Bin überlappen.

peak_bins$score <- countOverlaps(peak_bins, reads)

Funktion für Binned-Coverage

count_bins <- function(reads, target, bins){
  # Find all bins overlapping peaks
  overlap <- from(findOverlaps(bins, target))
  target_bins <- bins[overlap, ]

  # Count the number of reads overlapping each peak bin
  target_bins$score <- countOverlaps(target_bins, reads)
  target_bins
}

Coverage für Blacklist-Regionen

peak_bins <- count_bins(reads_ext, peaks, bins)
bl_bins <- count_bins(reads_ext, blacklist.hg19, bins)

Hintergrund-Coverage

Entferne alle bereits berücksichtigten Bins.

bkg_bins <- subset(bins, !bins %in% peak_bins & !bins %in% bl_bins)

Zähle die Reads, die jedes verbleibende Bin überlappen.

bkg_bins$score <- countOverlaps(bkg_bins, reads_ext)

Lass uns üben!

ChIP-seq mit Bioconductor in R