I want to plot a transcript with it's coverage data as a histogram using DataTrack, but I'm a bit stuck.. Consider this code:

library(Rsamtools)
library(RNAseqData.HNRNPC.bam.chr14)
library(TxDb.Hsapiens.UCSC.hg19.knownGene)

# Get reference to example bamfile
bamfile <-  RNAseqData.HNRNPC.bam.chr14_BAMFILES[1]
# Exmple transcript name
transcript <- "uc001yhh.1"

# Get exons for transcript
allExons <- GenomicFeatures::exons(
  TxDb.Hsapiens.UCSC.hg19.knownGene,
  vals = list(tx_name = transcript),
  columns = list("EXONNAME", "TXNAME", "GENEID"))
# Expand GRanges
allExons <- expand(allExons)
# Remove unwanted rows
allExons <- allExons[mcols(allExons)$TXNAME == transcript]
# Set $transcript mcols
mcols(allExons)$transcript <- mcols(allExons)$TXNAME
# Create track
trTrack <- Gviz::GeneRegionTrack(allExons)
# Plot
Gviz::plotTracks(trTrack)
# Transcript plot works well

# Get coverage
cov <- GenomicAlignments::coverage(bamfile, param = Rsamtools::ScanBamParam(which = allExons))
# Coverage only for my transcript
cov <- cov[allExons]
# Unlist
cov <- unlist(cov)
# Turn into numeric vector
cov <- as.numeric(cov)

# How do I create a DataTrack object of with these coverage values? I wish to plot the transcript with the coverage data I now have in cov.

Any ideas? I don't want to use an AlignmentsTrack. For this task I have to extract coverage data using coverage() with the bamfile.

Please check 

? DataTrack

for details on how to initialise objects of the class. I suspect that cov, before casting to numeric,  is an RLE object, so something along these lines should work:

dt <- DataTrack(start=start(cov), end=end(cov), data=runValue(cov))

Florian