geom_h5vc: geom_h5vc
In h5vc: Managing alignment tallies using a hdf5 backend

Description Usage Arguments Details Value Author(s) Examples

Plotting function that returns a ggplot2 layer representing the specified dataset for the specified samples in the region [positon - windowsize, position + windowsize].

1	geom_h5vc( data, sampledata, samples=sampledata$Sample, windowsize, position, dataset, ... )

`data`	The data to be plotted. Returned by `h5dapply`. Must be centered on `position`, extend by `windowsize` in each direction and contain a slot named like the `dataset` argument
`sampledata`	The sampledata for the cohort represented by `data`. Returned by `getSampleData`
`samples`	A character vector listing the names of samples to be plotted, defaults to all samples as described in `sampledata`
`windowsize`	Size of the window in which to plot on each side. The total interval that is plotted will be [position-windowsize,position+windowsize]
`position`	The position at which the plot shall be centered
`dataset`	The slot in the `data` argument that should be plotted
`...`	Paramteters to be passed to the internally used `geom_rect`, see `geom_rect` for details

Creates a ggplot layer centered on position using the specified dataset from list data, annotating it with sample information provided in the data.frame sampledata and showing all samples listed in sample. The resulting plot uses ggplot2's geom_rect to draw boxes representing the values from dataset. The x-axis is the position and will span the interval [positon - windowsize, position + windowsize]. The x-axis is centered at 0 and additional layers to be added to the plot should be centered at 0 also.

Ths function allows for fast creation of overview plots similar to mismatchPlot (without the stacking of tracks). The example below shows how one can create a plot showing the coverage and number of mismatches per position (but not the alternative allele) for a given region.

A ggplot layer object containing the plot of the specified dataset, this can be used like any other ggplot layer, i.e. it may be added to another plot.

Paul Pyl

  # loading library and example data
  library(h5vc)
  library(ggplot2)
  tallyFile <- system.file( "extdata", "example.tally.hfs5", package = "h5vcData" )
  sampleData <- getSampleData( tallyFile, "/ExampleStudy/16" )
  position <- 29979629
  windowsize <- 30
  samples <- sampleData$Sample[sampleData$Patient == "Patient8"]
  data <- h5dapply(
    filename = tallyFile,
    group = "/ExampleStudy/16",
    blocksize = windowsize * 3, #choose blocksize larger than range so that all needed data is collected as one block
    names = c("Coverages", "Counts", "Deletions"),
    range = c(position - windowsize, position + windowsize)
  )[[1]]
  # Summing up all mismatches irrespective of the alternative allele
  data$CountsAggregate = colSums(data$Counts)
  # Simple overview plot showing number of mismatches per position
  p <- ggplot() + 
  geom_h5vc( data=data, sampledata=sampleData, windowsize = 35, position = 500, dataset = "Coverages", fill = "gray" ) + 
  geom_h5vc( data=data, sampledata=sampleData, windowsize = 35, position = 500, dataset = "CountsAggregate", fill = "#D50000" ) + 
  facet_wrap( ~ Sample, ncol = 2 )
  print(p)